IMS Study Group Cantus Planus Papers read at the XVII meeting Venice, Italy 28 July - 1 August 2014 Ims Study Group Cantus Planus

Papers read at the xvii meeting Venice, Italy, 28 July - 1 August 2014 Ims Study Group Cantus Planus

Papers read at the xvii meeting Venice, Italy, 28 July - 1 August 2014 edited by James Borders

Edizioni Fondazione Levi Venezia 2020 FONDAZIONE UGO E OLGA LEVI Cantus Planus PER GLI STUDI MUSICALI ONLUS

Consiglio di Amministrazione Coordinamento editoriale IX Presentation Davide Croff Presidente Luisa Zanoncelli Davide Croff Luigi Brugnaro Paolo Costa Collaborazione editoriale XI Preface Fortunato Ortombina Lucia Boscolo (2, 3, 5, 10, 26) Roberto Calabretto - Luisa Zanoncelli Giovanni Giol Paola Dessì (6, 7, 17, 18, 31, 37) Nicola Greco Vicepresidente Alessandra Ignesti (15, 21, 22, 25) XIII Foreword to the Meeting Giancarlo Tomasin Nausica Morandi (28, 30, 34, 36, 38, 48) James Borders Silvia Tessari (43, 44, 45, 46, 47) Revisori dei Conti Claudia Canella XIV Opening address Presidente Raffaello Martelli Ilaria Campanella Nausica Morandi Chiara Boldrin Maurizio Messina Progetto grafico e impaginazione XVI Abbreviations Karin Pulejo Comitato scientifico Roberto Calabretto Presidente In copertina Sandro Cappelletto elaborazione grafica da I Notation Dinko Fabris Fondo Antico, Lat. Z. 506 (=1611) fol. 41r Laurent Feneyrou Biblioteca Nazionale Marciana, Venezia 3 Zur Bedeutung von Oxeia/Acutus/Virga in den griechischen Cormac Newark und lateinischen Neumenschriften Paolo Troncon Ringraziamenti Andreas Pfisterer Marco Tutino Biblioteca Nazionale Marciana, Venezia Paula Varanda Maurizio Messina 9 From France to northern Italy: specific features in ‘Comasca’ notation Vasco Zara Susy Marcon Laura Albiero Biblioteca San Francesco della Vigna, Venezia Direttore e direttore della Biblioteca Rino Sgarbossa 19 A methodology for studying Old Hispanic notation: some preliminary thoughts Giorgio Busetto Elsa De Luca Isola di San Servolo, Venezia Staff Fulvio Landillo Ilaria Campanella II Manuscripts, fragments, and sources Claudia Canella Alessandro Marinello 43 The ‘codification’ of new Latin song in early twelfth century: Fabio Naccari codicological insights into F-Pn fonds latin 1139 Anna Rosa Scarpa David Catalunya

Collaboratori 49 Manuscripts in Cortona: fragments and liturgical books Margherita Olivieri in the Archivio storico diocesano Camilla Zennaro Marco Gozzi Archivio Giovanni Morelli Paola Cossu 61 The Birgittine Abbey of Maria Refugie Laura Desideri Five hundred years of manuscript production Francesco Verona Karin Strinnholm Lagergren Valeria Zane © 2020 by FONDAZIONE LEVI Angelina Zhivova S. Marco 2893, Venezia 73 Liturgical fragments of the diocese of Sigüenza (eleventh-sixteenth centuries) Tutti i diritti riservati per tutti i paesi Santiago Ruiz Torres - Juan Pablo Rubio Sadia

edizione on-line 83 Imposing European culture on the Cape Colony: https://www.fondazionelevi.it/editoria/cantus-planus medieval manuscripts in the Grey collection ISBN 978 88 7552 058 8 Rebekka Sandmeier 95 A northern Italian intermediary between Avignon and Rome? 253 Textinterpretation und Formelkomposition. Heiligenoffizien im Codex Hartker Oxford, Bodleian Library, MS. Canonici Liturgical 375 Dirk van Betteray and the chants of the 1485 Pontificale Romanum James Borders 267 The articulation of direct speech in responsories Roman Hankeln 107 The antiphoner of Izola Jurij Snoj 279 Simeon: the long journey to Trier and Sainthood Danette Brink 117 Der Seckauer Liber ordinarius von ca. 1595 (A-Gu 1566) als letztes Dokument der mittelalterlichen Salzburger-Seckauer Liturgie und Musik Réka Miklós VII Regional and local traditions

293 The vespertini as witnesses to the transmission of Old Hispanic chant in traditions A and B III Editing medieval monophony: between the old and new philologies Raquel Rojo Carrillo

137 Towards a semiotically informed transcription practice 311 Hirsau in Norditalien: Zur liturgischen Musikpraxis Andreas Haug des Benediktinerklosters Moggio im zwölften und dreizehnten Jahrhundert Hanna Zühlke 143 Reconstructing acts of writing. Editorial consequences of writing scenarios assumed for the versus Annus novus in Paris 1139 323 The Office for Saint Cecilia in musico-liturgical manuscripts Konstantin Voigt from the Pontificia Biblioteca Antoniana of Padua Nausica Morandi 151 Working realities of the New Philology: considering the potential of technology in editing medieval texts 337 Antichi testimoni dell’ufficio di san Prospero e tradizione nei codici reggiani Elaine Stratton Hild Stefania Roncroffi

345 Il canto del Passio presso alcuni ordini religiosi in fonti italiane tardive IV Chant and computing Diego Toigo

157 Cantus for office and mass: building an online network of chant databases 359 Apostolicité et exports liturgiques en France méridionale, Gaule cisalpine Debra Lacoste - Jan Koláček et Italie non romaine (835-1150) Jean-François Goudesenne 163 A new way to see neumes: the optical neume recognition project in context and in action Kate Helsen - Inga Behrendt - Jennifer Bain 377 Le sanctoral de la cathédrale de Sens Océane Boudeau

V Alleluias, tropes, sequences, and the mass ordinary 389 The weekly commemorative office: a twelfth-century witness from Saint-Denis in context 171 Late flourishing of the alleluia repertory in central Europe Tova Leigh-Choate Gábor Kiss 401 The Visitatio sepulchri as a Gospel Harmony in medieval Germany 187 Tropes in cantus planus sources from South Tyrol Melanie Batoff Giulia Gabrielli 413 Singing the saints in Medieval Eichstätt: 197 Kyrie chants in manuscripts from the German-speaking area the case of Wolfhard of Herrieden’s office for Saint Walburga Marit Johanne Høye Alison Altstatt

213 Mensurally notated mass ordinaries from Cambrai and Tournai 425 Learning gender roles from the Paschal and Mariological processions Sarah Ann Long of the brothers and sisters at Vadstena Abbey Michelle Urberg 225 Agnus pairing and disappearing: a contribution to the late chant tradition in Bohemia Hana Vlhová-Wörner 437 Venetian influence in fifteenth-century Portugal Manuel Pedro Ferreira

VI Saints offices: new themes and contexts 451 The feasts of the Holy Virgin in the liturgy of the mass in Zagreb Cathedral and its mediterranean context 241 In search of black swans: Orsolya Csomó a computer-assisted approach to interval-pattern recognition Morné Bezuidenhout VIII Chant and liturgy in latin Southern Italy Davide Croff

465 Ferial office in the Montecassino antiphoner and in Beneventan sources of the eleventh and twelfth Centuries Presentation Thomas Forrest Kelly - Katarina Livljanić

481 The masses for the Dead in Beneventan manuscripts Issues of formulary organization and chant manipulation Luisa Nardini

495 Graphic difference and the interpretation of the climacus in Beneventan notation Matthew Peattie Ten years or so ago, professor Antonio Lovato, the then President of the Scientific Committee, who had just returned from the Vienna edition of an important 517 The hymn in Beneventan manuscripts Bibiana Gattozzi meeting of the International Musicological Society – Study Group ‘Cantus Planus’, recommended that, thanks to the Fondazione Ugo e Olga Levi, Venice should offer to host the following edition of the meeting, which had, until then, never been held IX Byzantium and the East in Italy. 533 The tropologion Vaticanus Graecus 771 Svetlana Kujumdzieva The Foundation was honoured to accept the proposal and to take on the organization 549 The Byzantine musical manuscripts of the Veneto region: of the event, which appeared to be very challenging given the number of participants. overview of a current research project. One example (Bassan. gr. 34B19) At the moment of celebrating the meeting, Luisa Zanoncelli, the President of the Silvia Tessari Scientific Committee of the Levi Foundation, guided the entire staff of the Foundation, 563 The Byzantine tradition of the great doxology which moved to the Laguna, on the island of San Servolo, a venue that was perfect Gerda Wolfram for hosting the event. Giorgio Busetto, Ilaria Campanella, Claudia Canella, Fabio 569 O quando in cruce – Ὅτε τῷ σταυρῷ revisited Naccari, Alberto Polo and Anna Rosa Scarpa, along with the community service Nina-Maria Wanek volunteers, Emilia Cervero and Filippo Gazzola, dedicated their time to the event, 579 The classification of Russian sticheraria: a preliminary case study organizing, among other things, the organ concerts by Letizia Butterin and the Svetlana Poliakova Ensemble Oktoechos, directed by Lanfranco Menga in San Servolo and San Francesco

585 A schematic musical form for singing Buddhist hymns della Vigna, where it was also possible to visit the important Franciscan Library in the ‘bilingual’ literary culture of early Japan and to witness the illustration of a number of codices delivered by Massimo Bisson. Elizabeth J. Markham Another very important moment was the visit to the monumental Sansoviniane halls of the Biblioteca Nazionale Marciana where, for the occasion, a display of Byzantine codices with musical notation was prepared, and of which the Foundation Levi edited 593 Cantus Planus 2014 Program of the meeting the catalogue. Sandra Martani guided the initiative, and Silvia Tessari prepared the labels. Since then, the Levi Foundation has worked closely with these researchers and has even just recently prepared important initiatives on Byzantine music. A collective volume edited by Tessari, which includes the proceedings of another conference held in the Foundation and is dedicated to the subject, is forthcoming.

We are now proud to announce that — after a long and difficult editing process, for which we must thank Luisa Zanoncelli, who tenaciously directed it, and those who collaborated with her in diverse manner: Lucia Boscolo, Paola Dessì, Alessandra Ignesti, Nausica Morandi, Silvia Tessari and Claudia Canella together with Ilaria Campanella — the volume of the Proceedings is now complete.

IX CANTUS PLANUS 2014

Thus again, the Fondazione Ugo e Olga Levi confirms the importance of its role Roberto Calabretto - Luisa Zanoncelli as a point of reference for the musicological studies of the Middle Ages and the Renaissance, while rapidly increasing its editorial production, which is now on-line Preface and open access.

Collected here are the proceedings of the seventeenth Ims Study Group Cantus Planus Conference, which was co-organized by the Department for Historical, Artistic, Musical and Demo-anthropologic Heritage of the University of Padua, and the Fondazione Ugo e Olga Levi in Venice, and which was held in Venice (Palazzo Giustinian Lolin / Isola di San Servolo) from July 28 - August 1. 2014. The Fondazione Levi is proud to have participated in the event as well as to have funded it: numerous scholars from three continents attended the event, which celebrated the thirtieth anniversary of the Study Group’s meetings (1984-2014), listed below:

Veszprém, Hungary, 1984 Budapest/Visegrád, Hungary, 2000 Bologna, Italy, 1987 Leuven, Belgium, 2002 Tihany, Hungary, 1988 Lillafüred, Hungary, 2004 Pécs, Hungary, 1990 Niederaltaich, Germany, 2006 Madrid, Spain, 1992 Zürich, Swiss Confederation, 2007 Eger, Hungary, 1993 Dobogóko˝, Hungary, 2009 Sopron, Hungary, 1995 Vienna, Austria, 2011 London, United Kingdom, 1997 Venice, Italy, 2014 Visegrád, Hungary, 1998

During the conference – which was divided into twenty-three sessions and two posters, an exhibit dedicated to Byzantin notation in the codices of the Biblioteca Marciana, and three concerts – the results of complex and original studies were presented. The proceedings – edited by James Borders with the support of the Fondazione Levi, and, for the musical examples, with the kindly advice offered to us by David Hiley and Marco Gozzi – demonstrate the depth of the Group’s projects,

X XI CANTUS PLANUS 2014

which have broadened the central themes during their thirty years of activity to James Borders, Chair, Ims Study Group «Cantus Planus» include new methodology and stimulating perspectives, also through a progressive branching off into active subgroups. Along with two of the subgroups (Historiae Foreword to the Meeting and Byzantin Music) the Fondazione Levi then organized two other conferences (2017, Historiae: Litugical chant for offices of the Saints in the Middle Age; 2018, Bessarione e la musica: concezioni, fonti teoriche e stili). The relative proceedings are in the process of being published and forthcoming. In confirming its readiness to cooperate further, the Fondazione Levi wishes to express its satisfaction for the superior scientific level of the papers published here. Founded in 1984 at the initiative of Professors Helmut Hucke and László Dobszay with the approval of the Directorium of the International Musicological Society (Ims), the Study Group «Cantus Planus» (Ims-Sgcp) has met regularly at international congresses of the Ims and independently every two or three years ever since. The objective of the Ims-Sgcp is the advancement of research in fields relevant to the history and practice of liturgical plainchant as branches of learning and scholarship. In pursuing this objective, the Ims-Sgcp encourages international cooperation and facilitates constructive interactions among researchers.

The Study Group held its seventeenth meeting on the small and peaceful island of San Servolo in the Venetian lagoon, once the site of a Benedictine monastery, between 28 July and 1 August 2014. This five-day gathering was co-sponsored by the Fondazione Ugo e Olga Levi and the Department for Historical, Artistic, Musical and Demo- anthropologic Heritage of the University of Padua. The organizers were Prof. James Borders, Ims-Sgcp chair, and Dr. Nausica Morandi, Università degli studi di Padova. The call for papers attracted over a hundred abstracts for sessions on individual topics, panel discussions, free papers, and posters on Western European and Eastern Mediterranean chant, as well as a session on polyphony. The program comprised twenty-three sessions, most of them simultaneous; many papers read at these sessions are published in this volume. Meeting participants also enjoyed a visit to a manuscript exhibition at the Biblioteca Nazionale Marciana near the Basilica di San Marco.

The Ims-Sgcp is most grateful to the Fondazione Ugo e Olga Levi for generously sponsoring both the Venice meeting and the publication of these proceedings. We extend special thanks to Luisa Zanoncelli, former Presidente del Comitato scientifico della Fondazione Levi, Giorgio Busetto, Direttore e direttore della Biblioteca Fondazione Levi, as well as the members of the Advisory Board, Christelle Cazaux-Kowalski, Zsuzsa Czagány, Lori Kruckenberg, Jeremy Llewellyn, Christian Troelsgaard, and Anna Vildera.

XII XIII CANTUS PLANUS 2014

Nausica Morandi

Opening address

Ladies and Gentlemen, kind colleagues, dear friends: it is with great pleasure that I welcome you to Venice, in the beautiful setting of the island of San Servolo, to the seventeenth meeting of «Cantus Planus». A special welcome to Prof. James Borders, chairman of the Study Group, and to the members of the Advisory Board.

I would like to thank the Fondazione Ugo e Olga Levi, and in particular the Board of Directors, the Director, dott. Giorgio Busetto, the Scientific Board, here represented by its President, Prof.ssa Luisa Zanoncelli, and the staff of the Foundation, for their precious and fundamental commitment to organizing this meeting. I also wish to thank the speakers and all the participants from twenty-six countries and four continents.

After the meeting of the Study Group in Dobogóko˝, Hungary, in 2009, and the one in Vienna in 2011, we are especially honored to host the Study Group in Italy for the first time under the prestigious patronage of Unesco. In these years of European and global crisis, organizing a scholarly meeting and investing financial, physical, and scientific resources to promote culture and research into chant present a particular challenge. Organizing this conference has been demanding work, which began in October 2012, but we have taken on that work with great enthusiasm and perseverance in the spirit of «Cantus Planus». This long work is balanced out today by the joy of seeing all of you here, in Venice, chosen by the Advisory Board on account of the city’s historical, cultural, and musical significance, and as a meeting place of West and East.

I hope that this meeting will provide many opportunities to enhance our knowledge and promote productive dialogue, discussion, and shared experiences, fostering cooperation among members and institutions, and the development of joint projects and new spheres of research, thus furthering the study of the medieval liturgical chant.

XIV XV CANTUS PLANUS 2014

Abbreviations

col./cols. column/columns I fol./fols. folio/folios n.d. no date Notation n. pr. no printer or. ed. original edition p./pp. page/pages s.s. sine siglum vol./vols. volume/volumes

Ah Analecta Hymnica

Cantus Cantus: A Database for Latin Ecclesiastical Chant- Inventories of Chant Sources. Directed by Debra Lacoste (2011-), Terence Bailey (1997-2010), and Ruth Steiner (1987-1996). Web developer, Jan Koláček (2011-). Available from

Cao Corpus Antiphonalium Officii-Ecclesiae

Cmm Corpus Mensurabilis Musicae < http://www.corpusmusicae.com/cmm.htm>

Grove The New Grove Dictionary of Music and Musicians, ed. Stanley Sadie, 20 vols. 1980

Grove2 The New Grove Dictionary of Music and Musicians, ed. Stanley Sadie and John Tyrrell, 29 vols., London-New York: Macmillian, 2001

Grove online Grove Music online, ed. Stanley Sadie < https://www.oxfordmusiconline.com/>

Jams Journal of the American Musicological Society

Mmb Monumenta Musicae Byzantinae

Mmmae monumenta Monodica Medii Aevi

PalMus Paléographie musicale

PL Patrologia Latina

Pmm Plainsong and Medieval Music

XVI 1 Andreas Pfisterer

Zur Bedeutung von Oxeia/Acutus/Virga in den griechischen und lateinischen Neumenschriften

Die Frage nach den Zusammenhängen zwischen unterschiedlichen Neumenschriften stellt sich vor allem auf zwei Ebenen, einerseits zwischen griechischen und lateini- schen Notationen, anderer­seits innerhalb­ der einen oder anderen Familie. Aus latei- nischer Sicht sind zwar die Zusammen­hänge innerhalb­ der östlichen Notationen we- niger relevant, der griechisch-lateinische Zusammenhang­ und die Gruppierung und Zuordnung der lateinischen Notationen lassen sich dagegen kaum von der Frage nach der Entstehung­ der musikalischen Notation im lateinischen Westen trennen. Für eine Abhängig­ ­keit der lateinischen Notationen­ von den griechischen hat Constantin Floros plädiert (1970, 2, pp. 5-237). Nachdem­ seine Darstel­ ­lung jedoch auf Widerspruch­ gestoßen war, ist die Diskussion­ ergebnis­los stehen­geblieben (Haas 1975). Zu den Verwandtschaften­ innerhalb­ der lateini­ ­schen Notationen­ gab es zwar schon mancher- lei Ansätze,­ aber keine ‘communis opinio’, in jüngere Zeit steht die Erforschung­ der Einzel­notationen so sehr im Vordergrund,­ dass Fragen nach den Zusammenhängen­ nur zaghaft gestellt werden.1 Dieser Beitrag­ versucht hierzu einen neuen Ansatzpunkt­ zu gewinnen und geht dabei von einem Zeichen aus, dass praktisch allen griechischen und lateinischen Neumen­schriften gemein­sam ist und jeweils zu den elementaren Bestand­teilen des Zeichensystems­ gehört.

Akzentschrift Als Leo Treitler vor gut dreißig Jahren einen Aufsatz zur Ent­stehung der musi­kali­ schen Schrift veröffentlichte,­ schien ihm die so genannte ‘Akzent­theorie’, wonach die Grund­zeichen der Neumen­schrift aus der alexandrinischen­ Akzentschrift­ stammen, so ab­wegig, dass sie nur noch einen Todesstoß­ bräuchte, um aus der Diskussion zu verschwinden­ (Treitler 1984). Sie verschwand aber nicht, im Gegen­teil: Mathias Bielitz (1989, pp. 79-115) und Charles Atkinson (2009, pp. 40-46, 54-65 und 101-113) haben deutlich gemacht, dass man nur die Akzentschrift­ richtig verstehen muss, um diese Theorie plausibel zu machen.

Dabei ist es hilfreich, sich die von Aristoxenos vorgenom­ ­mene Gegen­über­stellung von Sing- und Sprechstimme­ in Erinnerung zu rufen. Zur Sprechstimme­ gehört eine musi­kali­sche Raum­bewegung, die er zusammenhängend­ (συνεχής) nennt, die auf keiner Tonhöhe stehen bleibt; zur Singstimme­ gehört die abstands­ ­weise Bewegung

1. Zur aktuellen Diskussion um frühe lateinische Neumenschriften vgl. die Beiträge zweier Tagungen (Colette und Massip 2012; Aubert und Rankin 2013) sowie Rankin (2011 und 2014).

3 CANTUS PLANUS 2014 PFISTERER — ZUR BEDEUTUNG VON OXEIA/ACUTUS/VIRGA

(διαστηματική), die von einer stabilen Tonhöhe­ zur nächsten springt (Aristoxenus als die syllabische, kann man zumindest vermuten, dass die historische Entwicklung 1954, pp. 13-15). Da die Akzentschrift­ für die Sprechstimme­ konzipiert ist, gibt sie nicht von der melismatischen zur syllabischen Verwendung ging.2 Da aber Aufzeichnungen Tonhöhen­ an, sondern Bewegungs­richtungen, wobei die Null­bewegung, das Stehen­ melismatischer Stücke in archaischer Notation nicht erhalten sind, bleibt das Spe- bleiben auf einer Tonhö­ he, ausgeschlossen­ ist. Dem entsprechen­ auch die Zeichen kulation. Syllabische Stücke in archaischer Notation gibt es, sie zeigen zum Teil sehr Oxeia und Bareia, die in Schriftrichtung­ gelesen Aufwärts-­ und Abwärtsbewegung­ deutlich den stenographischen Charakter der byzantinischen Notation.3 Unwichtige dar­stellen. Solange­ man im Bereich­ der ge­sprochenen Sprache bleibt, ist aller­dings Silben bekommen überhaupt kein Zeichen, markiert werden neben Mehr­ton­gruppen eine genaue Trennung zwischen Aufwärtsbewegung­ und Hochton­ nicht notwendig;­ und rhythmischen Dehnungen die melodischen Hochpunkte,­ eben mit der Oxeia oder des­halb sind die Definitionen in der grammatischen Literatur häufig unscharf. Bei Petaste. In einer solchen Notation ist es schwierig zu sagen, welche Töne ein Zeichen der Über­tragung auf die Sing­stimme dagegen muss geklärt werden, was genau ein bezeichnet und welche es nur impliziert. Vielleicht ist das der Kontext, in dem das Zeichen bedeutet. Diese Fragen, die sich bei der Übertragung­ eines ursprüng­ ­lich Auseinander­ ­treten der syllabischen und melismatischen Verwendung der Oxeia zu sprach­lichen Zeichen­systems auf die Musik ergeben, scheinen mir zentral zu sein für verstehen ist. das Verständnis der Neumenschriften.­ Die Bareia sagt dagegen überhaupt nichts über das melodische Verhältnis zum Voran­ Byzantinische Notationen gehenden oder Folgenden, der Abstieg vollzieht sich innerhalb­ des Geltungsraumes­ In den älteren Stadien der byzantinischen Neumen­schrift unterscheidet­ man zwei des Zeichens. Dafür gibt es eine andere Besonderheit:­ Die zwei Töne stehen zwar Traditions­ ­stränge, diese haben aber einen gemein­samen Grund­bestand an Zeichen, zu meist auf einer Silbe, können aber auch auf zwei Silben verteilt werden, sogar über dem auch Oxeia und Bareia gehören; daher können die Unterschiede­ hier unberück­ Phrasenzäsuren­ hinweg (van Biezen 1968, pp. 26-28; Floros 1970, 1, pp. 143-144). sichtigt bleiben (Strunk 1955; Floros 1970, 1). Dies macht zumindest deutlich, dass die westliche Selbst­verständlichkeit,­ dass ein Zunächst fällt eine Asymmetrie auf: Die spiegel­bildlichen Zeichen Oxeia und Bar- Zeichen nicht über die Silbengrenze­ hinweggehen­ kann, so selbstver­ st­ ändlich nicht eia haben keine spiegel­bildliche Bedeutung. Die Oxeia steht für einen relativ hohen ist, sondern eine Grundent­ ­scheidung, die die lateinischen Neumenschriften­ unter­ Einzel­ton, die Bareia steht für zwei Töne in ab­steigender Richtung. Wie es zu dieser einander verbindet. Asymmetrie kommt, ist unklar. Das Notations­system macht den Eindruck,­ nicht von den Zeichen und ihrer ererbten Bedeutung her konstruiert zu sein, sondern von den Lateinische Notationen Bedürfnissen des Gegen­standes, der auf­ge­zeichnet werden soll. Für die lateinischen Neumenschriften­ ist eine von Leo Treitler gemachte Beobachtung entscheidend, die ich allerdings in eine andere Richtung interpretiere (Treitler 1982, Wie dem auch sei, die Bedeutung der Oxeia lässt sich relativ gut erschließen. In den pp. 250-254). Anders als im griechischen Osten gilt im lateinischen Westen die Grund- Lehrbüchern­ der mittelbyzantinischen­ Notation lernt man, dass sie für einen Sekund- regel, dass ein Notations­zeichen nur über das etwas aussagt, was inner­halb einer schritt aufwärts­ gegenüber dem vorangehenden­ Ton stehe und dass das Intervall Silbe passiert. Ich nenne das die ‘isolierende Silbengrenze’.­ Diese Grundregel­ wird durch zusätzliche Intervallzeichen­ modifiziert werden könne. Gegenüber dieser für dann – vermutlich sekundär – an verschiedenen Stellen durchbrochen,­ vor allem in die Intervallzeichen­ allein relevanten Beziehung zum vorangehenden­ Ton hat Jan van syllabischem Kontext. Nimmt man aber an, dass diese Grund­regel vor bzw. unter­halb Biezen (1968, pp. 17-20; vgl. Jammers 1962, pp. 49-50) klar gezeigt, dass im syllabi- der bekannten Gegen­belege gilt, dann ordnen sich die lateinischen Neumenschriften­ schen Kontext die Oxeia nicht höher sein muss als der voran­gehende Ton (möglich ist ganz plausibel. auch eine Ton­wieder­holung, aber offenbar kein Abstieg), aber immer höher ist als der folgende. Das muss also die ältere Bedeutung der Oxeia sein, die von der Regulierung Soll die Grundbedeutung­ der Aufwärts- bzw. Abwärtsbewegung­ unter den Bedingun- der Intervall­zeichen im zwölften Jahr­hundert überdeckt­ wird. Van Biezen (1968, pp. gen der isolierenden Silben­grenze bei­behalten werden, so muss die Bewegung inner­ 35-36, 38) konnte allerdings­ auch nicht leugnen, dass in Kombinationen mit einem halb der Silbe stattfinden,­ also steht der Acutus für eine zweitönige­ Figur. Genau das vorangehenden­ Zeichen auf derselben Silbe die Oxeia für einen Ton steht, der immer ist der Fall in der paläofränkischen­ Notation. Dort gibt es dann auch das Gegen­stück, höher ist als der vorangehende,­ aber nicht höher sein muss als der folgende. Es wird also dabei bleiben müssen, dass die Oxeia in syllabischer Verwendung auf den fol- 2. Denkbar wäre z.B., dass die Oxeia sich im syllabischen Kontext an die Petaste angepasst hat, die auch im melismatischen genden Kontext, in melismatischer Verwendung auf den voran­gehenden weist. Da die Kontext in der Regel einen Ton bezeichnet, der höher ist als der folgende. Van Biezen (1968, p. 38) führt hierzu nur eine Ausnahme an. melismatische Verwen­dung der Bedeutung des Akzent­zeichens ‘Aufstieg’ näher steht 3. Siehe etwa die synoptischen Beispiele bei Floros (1970, 3, pp. 167-210).

4 5 CANTUS PLANUS 2014 PFISTERER — ZUR BEDEUTUNG VON OXEIA/ACUTUS/VIRGA den Gravis, der für eine zwei­tönige absteigende­ Figur steht. Will man einen Einzel­ton je unter­schied­lichen Methoden erweitert, durch diastematische Anordnung im Raum auf einer Silbe schreiben, so braucht man ein Zeichen für die ‘Nullbewegung’;­ die- (vor allem bei Notationen, die nur ein Zeichen für den Einzelton haben), durch eine se Funktion wird sinnvollerweise­ von einem horizontalen Strich bzw. einem Punkt Systematisierung der Verwendung von Virga und Punctum (häufig mit Virga im Auf- über­nommen. Diese Zeichen sind in der paläo­fränkischen Notation als konsequente und Abstieg und Punctum am Tief­punkt, also gegenläufig­ zur Ver­wendung­ in Melis- Erweiterungen des Akzent-Systems erkennbar, nämlich um genau den Fall, den es in men) oder durch die Einführung­ eines dritten Zeichens zwischen Virga und Punctum. der Sprach­melodie nicht gibt.4 Gerade diese Versuche der Ver­besserung zeigen, dass der gemein­same Ausgangs­ ­punkt hier eine blinde Stelle hatte. In den so genannten Punktnotationen­ (bretonisch, aquitanisch, lothringisch) wird die Virga für einen Einzel­ton verwendet, der das Ergebnis und der End­punkt einer Betrachtet man die Bedeutung der Virga in dieser Weise, so wird zum einen ein Aufwärtsbewegung­ ist. Sonstige Einzeltöne­ werden mit Punkt bzw. waagrechtem Strich grund­legender Unterschied zwischen griechischer und lateinischer Notation in der bezeichnet. Die isolierende Silbengrenze­ führt nun aber dazu, dass die Virga nicht Bedeutung der Silben­grenze deutlich; zum andern rücken die drei Grund­typen der am Anfang einer Silbe stehen kann, sondern nur nach einem tieferen Ton. Deshalb lateinischen Notation in eine logische Reihenfolge, die nicht ohne weiteres umkehrbar wird zwar aufwärts­ Punkt-Punkt-Virga geschrieben, abwärts aber Punkt-Punkt-Punkt; ist. Die Verwendung von Punkt bzw. waag­rechtem Strich für den Einzel­ton ist in der ent­sprechend in allen anderen Kombinationen. Deshalb kann auch ein Einzelton­ auf paläo­fränkischen Notation als notwendige Ergänzung der Akzentzeichen­ erklärbar. einer Silbe nicht mit einer Virga geschrieben werden (Aus­nahmen in der bretonischen Im Rahmen der Punkt- wie der Akzentnotationen­ ist dagegen die Erklärung schwierig; Schrift mögen auf sekundärem Einfluss der französischen Notation beruhen). Eine die moderne Forschung hat sich hier lange Zeit mit der Annahme abgefunden, dass Symmetrie zwischen aufsteigenden­ und absteigenden­ Zeichen könnte man herstellen,­ Punkt bzw. Strich vom Gravis abgeleitet seien, ohne plausibel machen zu können, wa- wenn man den Gravis für den melodischen Tiefpunkt­ verwendete. Das wird aller­ rum dieses Zeichen seine Gestalt verä­ ndert habe.5 Akzeptiert man die hier vorgestellte dings selten gemacht, sei es nur, weil melodische Tiefpunkte weniger auffällig sind als Erklärung, dann sind Punkt- und Akzent­notationen von der paläo­fränkischen abhän- Hoch­punkte und daher keine besondere Markierung verlangen. gig und greifen nicht direkt auf das antike Akzentsystem­ zurück. Der unterschiedliche­ Umgang mit der Virga ist dann ein Indi­kator für eine Gewichtsverlagerung­ von der Die so genannten Akzent­notationen beseitigen diese Asymmetrie zwischen auf­ Notation von Bewegungen zur Notation von Einzel­tönen. steigenden und absteigenden Tongruppen; sie verwenden die Virga in allen Kontexten für den Hochton, geben dafür aber die ursprüngliche Bedeutung ‘aufsteigende Be- wegung’ preis. Das ist ein markanter Eingriff in die Zeichen­bedeutung. Der Vorteil ist, dass nun ein musikalisch wichtiger Punkt, der obere Umkehrpunkt­ der Bewegung, in allen Kontexten durch ein auffälliges Zeichen hervor­gehoben werden kann. Ein analoges Zeichen für den unteren Umkehrpunkt wird dagegen nicht verwendet, ob- wohl der Gravis dafür zur Verfügung stünde. In manchen Schriften wird der Gravis gelegentlich für den untersten Ton eines Abstiegs verwendet, das ist aber fakultativ und hat keinen festen Platz im System. Offenbar ist der Tiefpunkt­ musikalisch weniger interessant oder auffällig.

Von den Ausnahmen gegen die Regel der isolierenden Silbengrenze­ sei hier nur kurz die Notation in syllabischen Kontexten angeführt. In diesem Bereich gibt es zahlreiche­ regionale und chrono­logische Unterschiede.­ Vermutlich sind diese dadurch provo- ziert, dass Zeichen, die nicht über die Silbengrenze­ hinaus weisen, im syllabischen Kontext nur eine minimale Aussage­kraft haben. Diese Aussage­kraft wird dann mit

4. Gegen eine Ableitung des Tractulus vom gleich aussehenden antiken Zeichen für die Länge eines Vokals/einer Silbe (so Jammers 1965, p. 33 für die aquitanische Notation) spricht die fehlende Korrespondenz zwischen dem Punctum und dem halbrunden Kürze-Zeichen. Ebenso Floros (1970, 2, pp. 29-32). Gegen Floros’ Ableitung des Tractulus vom Oligon bzw. geraden Ison der byzantinischen Notationen (ibidem, pp. 32-35) spricht schon die Chronologie. Außerdem funktioniert die 5. Von den Autoren, die die Virga vom Accentus acutus ableiten, weichen nur wenige von dieser communis opinio ab: Ähnlichkeit der Zeichenbedeutung nur auf einer sehr vagen Ebene. Wagner (1912, p.116); Floros (1970, 2, pp. 29-35); Bielitz (1989, p. 106).

6 7 CANTUS PLANUS 2014

Laura Albiero Bibliography

Aristoxenus (1954), Elementa Harmonica, ed. Rosetta da Rios, Roma: Publica Officina Polygraphica. From France to northern Italy:

Atkinson Charles (2009), The Critical Nexus: Tone-System, Mode, and Notation in Early Medieval Music, specific features in ‘Comasca’ notation Oxford: Oxford University Press.

Aubert Eduardo Henrik - Rankin Susan eds. (2013), Actes du colloque d’Auxerre (17-18 juin 2011: Notarum figura: l’écriture musicale et le monde des signes au IXe siècle), Solesmes: Éditions de Solesmes (Études grégoriennes 40). The development and the spread of Metz notation is not deeply studied, if compared Bielitz Mathias (1989), Die Neumen in Otfrids Evangelienharmonie: Zum Verhältnis von Geistlich und with other more explored ones as the Beneventan or the Saint Gall notation (see Le Weltlich in der Musik des frühen Mittelalters sowie zur Entstehung der raumanalogen Notenschrift, Heidelberg: Universitätsbibliothek. Codex 339; Le Codex 10673; Le Codex VI.34; Kelly 1992; Livljanic 2014). Indeed, the Messine notation was one of the most important family of early musical script, and it Biezen Jan van (1968), The Middle Byzantine Kanon-Notation of Manuscript H: A Paleographic Study with a Transcription of the Melodies of 13 Kanons and a Triodion, Bilthoven: Creyghton. surely deserves a complete and exhaustive analysis (see Antiphonale missarum Sancti Gregorii). Colette Marie-Noël - Massip Catherine eds. (2012), Actes du colloque de Royaumont (29-31 octobre 2010): Manuscrits notés en neumes en Occident, Solesmes: Éditions de Solesmes (Études grégoriennes 39). As is well known, the Messine notation was used from the tenth century in a specific Floros Constantin (1970), Universale Neumenkunde, 3 vols., Kassel: Bärenreiter, 1970. region in north-eastern France, and commonly written in important cities like Reims,

Haas Max (1975), Probleme einer «universalen Neumenkunde» in Basler Studien zur Musikgeschichte 1, Bern: Laon, Arras, Verdun and so on. Francke (Forum Musicologicum 1), pp. 305-322. The first attempt to draw up an exhaustive list of sources in Messine notation was Jammers Ewald (1962), Musik in Byzanz, im päpstlichen Rom und im Frankenreich: Der Choral als Musik der Textaussprache, Heidelberg: Winter. made by Jacques Hourlier (1951). He found at least fifteen centers in which the Messine notation was used, mostly in north-eastern France and in Belgium. He noted – (1965) Tafeln zur Neumenschrift, Tutzing: Schneider. the spread of this notation in Bohemia and in Italy, giving for the latter one a list of Rankin Susan (2011), ‘On the Treatment of Pitch in Early Music Writing’, Early Music History 30, pp. 105- fourteen sources, some of them complete, some others fragmentary.1 175.

– (2014) Identity and Diversity: The Idea of Regional Musical Notations in Frank Hentschel - Marie Actually, this musical script was used in northern Italy, in a small region between Winkelmüller eds., ‘Nationes’, ‘Gentes’ und die Musik im Mittelalter, Berlin-Boston: De Gruyter, Milan and Switzerland, where the most important center was the city of Como. This pp. 375-393. region is not side by side with the French Messine area, but in some way this notation Strunk Oliver (1955), ‘The Notation of the Chartres Fragment’, Annales musicologiques 3, pp. 7-37 (reprint: arrived there and was used for at least two centuries and even more. It appears at the Essays on Music in the Byzantine World, New York: Norton, 1977, pp. 68-111). beginning in its adiatematic form, then it is written on lines.

Treitler Leo (1982), ‘The Early History of Music Writing in the West’, Journal of the American Musicological Society 35, pp. 237-279 (reprint: Treitler 2003, pp. 317-364). The Italian version of the Messine notation shows specific features which make this type

– (1984), ‘Reading and Singing: On the Genesis of Occidental Music Writing’, Early Music History 4, immediately recognizable: for that, this particular type is called ‘Comasca’ notation. pp. 135-208 (reprint: Treitler 2003, pp. 365-428). Bannister (1913, pp. 99-100) recognized as first the peculiarity of this notation. One of the most characteristic neumes is the tractulus, which has the shape of a hook (figure – (2003), With Voice and Pen: Coming to Know Medieval Song and How it was Made, Oxford: Oxford University Press. 1, p. 10): we can find this sign in some French manuscripts too, but it became the only type used in the Italian Messine sources. For this reason, it is quite easy to gather the Wagner Peter (19122), Einführung in die Gregorianischen Melodien: Ein Handbuch der Choralwissenschaft II: Neumenkunde: Paläographie des liturgischen Gesanges, Leipzig: Breitkopf und Härtel. sources. On the other hand, what makes this research difficult is the fact that most of the manuscripts were destroyed or lost, or they survive in a fragmentary form.

1. Hourlier’s list of Italian sources includes: I-Rn Sessoriano 136; I-VCd 186; I-CHV w.s.; I-Ma E 68 sup.; I-NOVasd w.s.; V-CVbav Reg. lat. 1553 (fols. 36-37); I-Ma E 72 inf.; V-CVbav Vat. lat. 82; I-MC 494; I-Ma M 70 sup.; F-Pnm n.a.l. 1410, 1411, 1412 and 1413.

8 9 CANTUS PLANUS 2014 ALBIERO — FROM FRANCE TO NORTHERN ITALY: SPECIFIC FEATURES IN ‘COMASCA’ NOTATION

manuscripts, containing respectively the winter and the summer part of the liturgical year. The sanctoral has no particular evidence of its origins, but according to a note written by Antonio Olgiati, the first prefect of the Ambrosian Library, it comes from Figure 1. Tractulus of the Comasca notation Olivone in Val Blenia, in southern Switzerland. Nevertheless, it’s quite unlikely that the manuscript was copied or used there, because in the eleventh century Olivone was a part of the diocese of Milan and followed the Ambrosian liturgy. The use of Messine notation in Italy began with the foundation of the monastery of Sant’Abbondio in Como, erected in 1010 by the bishop Alberic with the endorsement I-CHV without shelf-mark (Ruini 1990), antiphonary of the secular cursus, probably of the imperial court (Tatti 1683, 1, pp. 828-833). The sources are not clear about the from Chiavenna itself, of the first half of the eleventh century. It is made up by two provenance of the monks: according to Saverio Xeres (1996, p. 43), they came from parts originally independent and then put together; the first hand displays the French the monastery of Pfäfers, in Switzerland, because the ledger of the community of Messine notation, while the second has the Comasca notation. The compresence in Pfäfers quotes five «nomina fratrum de monasterio Sancti Abundi», which is indeed the same manuscript of the two types of Messine notation is an argument in support the monastery of Como. Nevertheless, this quotation is later than a list drawn in the of a French provenance of the monks who founded the monastery of Sant’Abbondio. third quarter of the eleventh century and, even if it shows some relation between the I-Rn Sessoriano 136, a ritual with main masses of the temporal and the office of the two monasteries, it does not mean necessarily that the monks of Sant’Abbondio came dead, of the second quarter of the century. This is probably from Sant’Abbondio of from Pfäfers. Como, because in the litanies there is the invocation ‘ut cunctam congregationem nostram sancti Abundi conservare digneris’; On the contrary, we can observe that in Pfäfers the musical notation was the Saint-Gall one, and, if the monks of Pfäfers went to Como, they would have brought their own I-MC 494, a hagiographical collection on Saint Remigius of Reims with a secular books and would have kept on writing their own notation. The only fact that in the office at the end, dated from the third quarter of the eleventh century. It was assigned Como region the Messine notation was used is enough to suggest the provenance of to Berceto (Hourlier 1954), not far from Parma, because it contains the story of a the monks from north-eastern France. miracle occurred there and related to the foundation of the church of Sant’Abbondio in Berceto. But in my opinion this is not enough to establish its origin, and actually I The use of the Messine notation does not exactly coincide with the diocese of Como: think that this manuscript is from Como too. on one hand, the Messine notation was not the only one used in this region, for we can find in some border areas, like Bormio and Chiavenna, some fragments in Saint-Gall The fragments are from missals and antiphonaries. They were re-used most of all notation (Rainoldi and Pezzola 2003, no. 10); on the other hand, we find Messine notation as flyleaves or binding pieces for notarial documents or printed books and are now in sources kept in some villages outside the Como region, like Boccioleto in Valsesia kept in Libraries and Archives in Milan, Pavia, Sondrio, Chiavenna, Como, Novara, (Baroffio 1966). Since the foundation of Sant’Abbondio is indeed the starting point of Lugano and in the Vatican Library. An exhaustive list of manuscripts and fragments Comasca notation, we will consider here only eleventh-century sources. For this period, in Comasca notation is given by Giacomo Baroffio (2011, p. 110, n. 32). The list of only five manuscripts survived; besides those, we have thirty-two known fragments. fragments is going to increase, because some new Comasca fragments have been recently uncovered in the examination of some northern Italian archives, most of all The five manuscripts are: the State Archive of Pavia and Milan: the latter one, for example, has not catalogued all the fragments yet.2 I-VCd 186 (Cereghetti 2006), a gradual of the end of the eleventh century. It comes from a church dedicated to saint Victor, because this saint, as well as saint Abundius, One of the most remarkable features of this notation is the use of some brachygraphical is marked in capital letters in the litanies and his mass has tropes; it is probably from signs – I prefer here to use the term ‘brachygraphy’ instead of ‘tachygraphy’ already the church of Saint Victor in Balerna, in the Swiss Canton of Ticino, but recently employed, because the purpose of the notator was not to write faster but to give a some other scholars assigned this manuscript to the church of Saint Victor in Como. significant sign to avoid copying a long series of neumes. By brachygraphical signs

I-Ma E 68 sup. (Sesini 1932), a gradual of the second quarter of the eleventh century. 2. They are about five thousand fragments, among which more than a half are liturgical. I’d like to thank Davide Dozio for Written by two different hands, it’s probably the result of the gathering of two original having let me consult the fragments there.

10 11 CANTUS PLANUS 2014 ALBIERO — FROM FRANCE TO NORTHERN ITALY: SPECIFIC FEATURES IN ‘COMASCA’ NOTATION

I mean some special ‘neumes’ which stand for an entire melodic formula. Two of these Graduals in mode 1 signs are used to indicate a melodic cadence, in particular a melodic formula which D1: [GF aGa aaF aGaFD GFF FEC] DDCDA CDFFDCD FFGFF DED stands on the last syllable but one, followed by a final clivis: the former, which has the shape of a ‘fermata’, refers to the scandicus flexus resupinus (figure 2); the latter, D2: [abGF accaG aGaFEFG] DEFG FaGFED EFD which has the same shape but horizontally inverted, is used to indicate the scandicus subbipunctis resupinus (figure 3). The use of these signs is not systematic, for we Graduals in mode 2 find also the formulas with traditional neumes, and sometimes different sources use them in different points of the same chant.3 More problematic is the third sign: it has a1: [cacd dcdedc] ece fdb caG abcd cde dedeca bca the shape of a quarter’s pause (figure 4) and it is not used to mean always the same d1: [edcb caG] bGa cc dede cefdb caG abaGa defe fede [ed] melodic line. It indicates, in general, a melisma held in one syllable, that is to say the a2: [ec edccac cac] cca cdedc edcda [a] last syllable of a chant or the last syllable of a musical phrase, but not always the same melisma. a3: [cacd decbcd] Gac cedcba bca a4: [cbcdecb ccb] dcdca [a]

a5: [dcbG acbG] ac bdeca

F1: [acaG] acca ccba cGGF Figure 2 Figure 3 Figure 4 Brachygraphical sign for the Brachygraphical sign for the Brachygraphical sign scandicus flexus resupinus scandicus subbipunctis resupinus for the melismas Tracts in mode 2

D3: [GFGFDE FaGa FED] DCDFF GaGEF EFaGF FED EED The first scholar who noticed this particular sign was Ugo Sesini (1932, p. 20). He published a short study on the gradual of the Ambrosian Library in which he explained Graduals in mode 5 this sign as an abbreviation for ut supra, as above. According to Sesini, the notator used F2: [acba ccc cca babGF] abGF aG ccc caa FaG aGGF this sign to avoid the writing of the entire formula and to indicate that the melody is the same as the melisma already given before. But if we check all the cases in which this F3: [dcca ccdech ccca ccca cca] FGa ccde hca GbaF sign is used, we see that this sign appears even if the neumatic formula is not entirely F4: [GF FFa GbaF] FaG ccc caa FaG aGGF written before, and, most of all, it appears in relation with some specific melismas. F5: [FEFGaG FGacaG] FGa cc GbaF

The examination of the places in which this sign appears has revealed that this is a6: [ccdccG cdcdccG] cdfgfd eca chdec edc edc cha linked to a particular method of composition which uses centonic units: the ‘composer’ c1: [acdcdc daF acdaF acab] acdcdc daF GF FFF acdec takes some melodic formulas which he arranges, according to the mode and the text, in order to make a new melody. This is the case, already studied, of some graduals in Graduals in mode 7 the fifth mode. G1: [Gcc chcdcd ddh dfed feddh] cdhcaG ahaaG This sign appears only in two Comasca manuscripts, the one of Vercelli and the one G2: [Gd dG dhdefd efed] ec ddh cdh caG aFG ccaG aFD Gca haaG of the Ambrosian Library. It concerns only some mass chants, and in particular it is used in graduals, tracts and alleluia verses. The melismas in which the sign is used G3: [Gahcd dedc chc dedcchd] fffc dedh cdh caG ahaaG have been classified according to the final note of the cadence: a progressive number G4: [Gah hdch eddh chcaG] hcaG ha ddd dhh Gha haaG has added in order to distinguish different formulas ending on the same note; the parts inside brackets indicate the previous and following sections, noted in neumes. Tracts in mode 8

G5: [chcaG GGaG GFG] ccc dcha aG achaG GaF ahcha haaG 3. On the brachygraphical sign for the scandicus subbipunctis resupinus see Albiero (2014).

12 13 CANTUS PLANUS 2014 ALBIERO — FROM FRANCE TO NORTHERN ITALY: SPECIFIC FEATURES IN ‘COMASCA’ NOTATION

Actually, we see that the melismas are specific not only in relation with the mode, Graduals in mode 5 but also with the typology of chant. I have found only one exception to this rule: the fifth-mode alleluia Beatus vir qui timet has, in the final syllable of the verse, the same Adiuvabit … F2 melisma used in the graduals in fifth mode (cadence F2). Bonum est confidere … F2

It has to be noticed that the first part of the melisma is usually notated with the ordinary Discerne causam … F2 neumes. This is a common practice: the neumes indicate the beginning of the formula, Iustus non conturbabitur … F2 which has to be recognizable thanks to the first neumes. Not all the melismas display the brachygraphic sign: in some chants the same melisma is entirely notated with the Priusquam te formarem … F2 proper neumes, for example in the gradual Bonum est confiteri, final cadence. In a few Qui operatus est … F2 cases, only the first neumes of the melisma are given, but they are not followed by the brachygraphic sign. Sederunt principes … F2

Suscepimus deus … F2 The case of the tracts – in the second and eighth mode – is the simplest one: the sign is given in the melisma that appears in the last syllable of the chant. The melisma with Timebunt gentes … F2 cadence in D is always the same for all the tracts in second mode, while the tracts in the eighth mode display their own melisma with cadence in G. Tollite hostias … F2 Iustorum animae … c1 … F2 Tracts in mode 2 Ecce sacerdos magnus … a6 … F2

Domine non … D3 Exiit sermo … a6 … F2

Domine audivi … D3 Diffusa est gratia … F3

Bonum est confiteri … F3 … F3(n) Tracts in mode 8 Convertere … F4 … F2 Qui seminant … G5 Esto mihi … F3(n) … F2 Beatus vir … G5 Propter veritatem … … F5 Commovisti … G5

Iubilate domino … G5 The four graduals in the seventh mode which have the brachygraphic sign feature Vinea facta est … G5 every time a different melisma.

Attende caelum … G5 Graduals in mode 7 Sicut cervus desiderat … G5 Dirigatur … G1

Miserere mihi domine … G2 The same technique appears in the graduals in the fifth mode: for the most part of them, only the final cadence in F has a melisma which uses the brachygraphic sign; Salvum fac … G3 only six graduals use the sign in an internal melisma with cadence on F or a or c (the Clamaverunt iusti … G4 ‘n’ indicates the cadence written with proper neumes).

14 15 CANTUS PLANUS 2014 ALBIERO — FROM FRANCE TO NORTHERN ITALY: SPECIFIC FEATURES IN ‘COMASCA’ NOTATION

The case of the second-mode graduals is the more interesting one. In them we find a massive use of the brachygraphic sign. We know that a group of graduals in the second mode is based on a fixed melody which is the same for different texts, and that the cadence on a is a reminiscence of an archaic form. In this group of graduals the Figure 5. Brachygraphical sign for the melismas in French Messine notation abbreviation sign is used in fixed melodic lines given on the last syllable of a phrase, and most of all in the last syllable of the verse. The presence of these types of sign concerns, more largely, the meaning of the use of Graduals in mode 2 neumatic notation: since the adiastematic notation does not show the intervals between notes, it is not useful to retrace the melody; on the other hand, if the singers knew the Hodie scietis a1 … d1 … a2 … F1 … a3 melodies by heart, they would not need to write them. The knowledge of liturgical Tecum principium a4F1 … a1 … d1 … a2 … a5+F1 … a3 music was certainly in the middle between the two cases: they did know the melodies, Iustus ut palma a4F1 … a1 … d1 … a2 … a5+F1 … a1 and they did learn by heart some melodic patterns, but they also needed a guide to recall which melody was to be sung in that particular text. The use of brachygraphical Domine refugium a4F1 … a1 … d1 … a2 … a5+F1 … a1 signs could confirm this hypothesis: in fact, once the first neumes are given, it was Angelis suis a4F1 … a1 … d1 … a2 … … a3 easy to sing the entire melisma which was known by heart.

Ab occultis a4F1 … a1 … d1 … a2 … a5+F1 … a3 The use of brachygraphical signs concerns also the writing process of these Ne avertas a4F1 … a1 … d1 … … … a3 manuscripts. Since it is indeed a very specific practice it is quite likely that all the

Haec dies … Confitemini a4F1 … a1 … … … a4+F1 … a3 Comasca manuscripts were copied in the same scriptorium. The diocese of Como was not of great extent, and only one working scriptorium was enough to cover the overall Haec dies … Dicat [a4F1] … [a1] … … … a4+F1 … a3 need of liturgical books. Besides, the five surviving manuscripts reveal the same Haec dies … Dicant [a4F1] … [a1] … … a2 … a5+F1 … a3 codicological features: they are small books, quite poor in the decoration and, most of all, they show the same ruling system. This last feature could not be very specific; but Haec dies … Dextera [a4F1] … [a1] … d1 … a2 … a5+F1 … a3 in the eleventh century the most common ruling system was the so-called ‘new style’, with dry point ruling made on every bifolium; the Comasca manuscripts have a ruling system called ‘two old style’ (Rand 1927), with two opened bifolia ruled together, The composition of these graduals uses centonic formulas to build up the chant: every which is quite uncommon in the eleventh century. All these elements suggest that all centonic section is introduced by a standard group of neumes which suggests the manuscripts come from the same scriptorium, probably the one of Sant’Abbondio in following melisma. This technique is well known among the responsorial chants of Como. the mass, and the fact that these melismas were learn by heart could explain the fact that the abbreviation sign is enough to indicate the whole melody: in fact, the reader The present contribution is only a first step that intends to reconstruct the musical was perfectly able to reconstruct the melody thanks to the first written neumes. written world of the Comasca region. The history of this type of neumatic notation is only a small part of musical and liturgical history in medieval northern Italy (Baroffio The Comasca notation is not the only one that used a conventional sign to mean a whole 2009; 2011). As we all know, many types of notations were used, every one linked to a fixed melodic section: in fact, for the melismas held in the last syllable of a chant, even small center: we hope that drawing out this little piece of the puzzle will help to gain a the French Messine notation used a kind of abbreviated sign, as we can clearly see in more wider and fuller understanding of these particular phenomena. the Gradual of F-LA 239 (figure 5, p. 17). The specificity of Comasca notation is the use, more or less systematic, of this sign for every centonic section, not only at the end but even inside the chant. Moreover, the Comasca notation applied the use of conventional signs to the other two formulas mentioned above, the scandicus subbipunctis resupinus and the scandicus flexux resupinus: this fact has no correspondence in the French Messine notation, nor in any other types of neumatic notation.

16 17 CANTUS PLANUS 2014

Elsa De Luca Bibliography

Albiero Laura (2014), ‘La tachygraphie musicale dans les sources messines-comâques: le scandicus A methodology for studying Old Hispanic notation: subbipunctis resupinus’, Études grégoriennes 61, pp. 37-63. some preliminary thoughts – (2016), Le fonti liturgico-musicali della diocesi di Como (sec. XI), Lugano: Vox Antiqua (Monographiae 1).

Antiphonale missarum Sancti Gregorii, IXe-Xe siècle, codex 239 de la Bibliothèque de Laon (19922), Solesmes: Abbaye de Solesmes (PalMus 10).

Bannister Henri Marriott (1913), Monumenti Vaticani di paleografia musicale latina raccolti ed illustrati, The beginning of Catholicism in Iberian Peninsula is generally dated to the Council Leipzig: Harrassowitz (Codices e vaticanis selecti phototypice expressi 12). of Toledo of 589, when Arianism was rejected.1 Among the various pre-Gregorian Baroffio Giacomo (1966), ‘Il Messale di Boccioleto’, Rivista di Storia della Chiesa in Italia 20, pp. 34-43. liturgies, the Old Hispanic liturgy is the one which held out the longest against the

– (2009), ‘Notazioni neumatiche (secoli IX-XIII) nell’Italia settentrionale: inventario sommario’, Aevum imposition of the Gregorian rite; however, in the Hispanic March the Franco-Roman 83, pp. 529-579. rite had already replaced the local liturgy from the eighth century (Gutierrez 2013,

– (2011), Music writing styles in medieval Italy in John Haines ed., The Calligraphy of Medieval Music, pp. 547-548). The Old Hispanic and Gregorian liturgies co-existed for some centuries Turnhout: Brepols, pp. 101-124. in the Iberian Peninsula before the Old Hispanic liturgy was eventually abandoned at the Council of Burgos of 1080, when the adoption of the Franco-Roman rite was Cereghetti Mirta (2006), Il graduale comasco Bibl. cap. Vercelli CLXXXVI, Master Dissertation in Musicology, Pavia University. finally formalized.2 Within the larger picture of Western sacred music, Old Hispanic chant is the pre-Gregorian repertory which is preserved most completely and has few Hourlier Jacques (1951), ‘Le domaine de la notation messine’, Revue Grégorienne 30, pp. 96-113; 150-158. Gregorian contaminations (Randel 1985, p. 230). Its importance for chant studies lies – (1954), ‘L’extension du culte de saint Remi en Italie’, Études Grégoriennes 1, pp. 181-185. in the fact that its study may unveil important information about Western liturgical

Kelly Thomas Forrest (1992), Les témoins manuscrits du chant bénéventain, Solesmes: Abbaye Saint-Pierre chant before the Carolingian reform which is otherwise lost. (PalMus 21). Today there are some forty surviving liturgical Old Hispanic manuscripts with music, Le Codex 339 de la Bibliothèque de Saint-Gall (Xe siècle) (1889), Solesmes: Abbaye de Saint Pierre (PalMus I/1). fragments included (Randel 1973; Zapke 2007, pp. 250-427). There are two types of Le Codex 10673 de la Bibliothèque Vaticane, fonds latin (XIe siècle). Graduel Bénéventain (1931), Tournay: Old Hispanic notation usually distinguished according to the ductus of the neumes on Soc. Saint-Jean l’Evangéliste - Desclée & Cie (PalMus 14). the page; the so-called ‘vertical notation’ is found in manuscripts from Northern Spain Le Codex VI.34 de la Bibliothèque Capitulaire de Bénévent (XIe-XIIe siècle). Graduel de Bénévent avec Prosaire and in three Toledan manuscripts; while ‘horizontal notation’ is found exclusively in et Tropaire (1937), Tournay: Soc. Saint-Jean l’Evangéliste - Desclée & Cie (PalMus 15). Toledan manuscripts (Brou 1955, pp. 29-31; Zapke 2011). Old Hispanic notation falls LivljaniĆ Katarina (2014), Montecassino, Archivio dell’abbazia, MS. 542: antiphonaire (XIIe siecle), Solesmes: within the family of ‘Frankish neumes’, broadly defined; notation is writtena campo Abbaye Saint-Pierre (PalMus 23). aperto, that is, the neumes are placed over the text and there are no dry point guide lines. Rainoldi Felice - Pezzola Rita (2003), ‘Apes debemus imitari. Ricerca sui frammenti liturgici della chiesa di Como (II)’, Archivio Storico della diocesi di Como 14, pp. 11-92. This paper presents the first results of my investigation into Old Hispanic notation carried out at the University of Bristol Rand Edward Kennard (1927), ‘How many leaves at a time?’, Palaeographia latina 5, pp. 52-78. as Postdoctoral Research Assistant in the Erc-funded Research Project ‘Shaping Text, Shaping Melody, Shaping Experience in and through the Old Hispanic Office’ (grant 313133). Further developments of this research are currently part of a work in progress (De Luca, n.d.). I wish to thank Prof. Susan K. Rankin for her comments and feedback on this research, Ruini Cesarino (1990), Particolari noti e poco noti di un Antifonario di Chiavenna dell’Undicesimo secolo in generously provided on several occasions. In addition, this research would have not been possible without the possibility of Dinko Fabris - Anselmo Susca eds., Tradizione manoscritta e pratica musicale. I codici di Puglia. Atti studying the Antiphoner in the Archive of León Cathedral. I am most grateful to Don Manuel Pérez Recio, Archivist of León del Convegno di Studi, Bari, 30-31 ottobre 1986, Firenze: Olschki, pp. 99-118. Cathedral, for granting access to the manuscript and for his hospitality in the Archive.

1. Useful overviews on the origins, adoption and development of Catholicism in the Iberian Peninsula, and the preparation Sesini Ugo (1932), La notazione comasca, Milano: Casa editrice d’arte e liturgia Beato Angelico. of the first liturgical books, can be found in Levy (1984 and 1987); Asensio Palacios (2008, pp. 135-143); Hornby and Maloy (2013, pp. 1-27); and Ferreira (2016, pp. 4-5). On the phenomenon of ‘Mozarabism’ see in particular Asensio Palacios (2008, Tatti Primo Luigi (1683), Degli Annali sacri della città di Como […], 4 vols., Milano: Giovan Battista Ferrario, p. 139). vol. 2. 2. For a list of the monasteries that abandoned the Old Hispanic rite for the Franco-Roman one during the twenty years before the Council of Burgos see Asensio Palacios (2008, pp. 146-147). While the replacement of the Old Hispanic rite with Xeres Saverio (1996), La fondazione monastica di Sant’Abbondio e il rinnovamento religioso in diocesi nel the Roman rite was not immediate, by the early twelfth century, only a few parishes in Toledo continued to follow the Old secolo XI in Carlo Calori - Dante Lanfranconi - Giuliano Signorelli - Giovanni Spinelli - Saverio Hispanic rite. On the liturgical reform launched in 1080 see, inter alia: Fernández de la Cuesta (1985b); Gonzálvez (1985); Xeres eds., La basilica di Sant’Abbondio in Como. IX Centenario della consacrazione 1095-1995, Como: Ruiz (1985); Rodríguez Suso (1992); Nelson (1996, pp. 12-15); Walker (1998); Vones (2007); Rubio Sadia (2011); Pick (2013); Il Settimanale - Società archeologica comense, pp. 35-51. Henriet (2017).

18 19 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

The inclination of the pen-stroke simply indicates whether the melody goes up or down on Maundy Thursday. The Franco-Roman Gradual of Gaillac F-Pn lat. 7767 contains and does not indicate the size of the intervals (Hornby and Maloy 2013, pp. 315-326). The some preces in Aquitanian neumes which can be also found notated in Old Hispanic lack of pitched notation means that decoding the notational symbols presents a far more neumes in E-Mh Cod. 56 (Aemilianensis 56, online) and E-SI ms. 4. To this group of complex challenge; as a matter of fact, chants in the Old Hispanic manuscripts cannot be Old Hispanic melodies in diastematic notation Rodriguez Suso added six pieces for the transcribed into modern notation. Modern palaeographers usually study chants in non- ordo of the dedication of the altar found in some Pontificals from Narbonne and other pitch readable notation by comparing them with their later versions written by more manuscripts whose melodic contour match the neumes found in the León Antiphoner recent hands using a pitched notational system. For the Old Hispanic repertory, however, for the same chants (Rodriguez Suso 1998 and 2004 pp. 95-99). Germán Prado (1934) apart from a few chants – about twenty-five out of a repertory of a few thousand – there focused on the music for some Lamentations found in E-SI ms. 9 from the Galician are no surviving later versions which could give any hint of the melodic content. monastery of San Rosendo de Celanova; these melodies show unusual characteristics for the Franco-Roman repertory, sharing instead similarities with some melodies in Despite the obstacle of the non-pitched notation and the peculiar history of the Old Old Hispanic neumes found in some earlier Bibles (Asensio Palacios 2009, pp. 20-21). Hispanic manuscripts with respect to the development of music writing, scholars who turned their attention to Old Hispanic chant managed to clarify some important aspects The Old Hispanic melodies written in diastematic notation certainly provide some of its musical tradition. Old Hispanic manuscripts received great attention in the 1980s clues toward a better understanding of the Old Hispanic music, but this set of chants with many articles and essays published in 1985 and an important congress held in represents a small percentage in a repertory of a few thousand Old Hispanic melodies. Salamanca in the same year. Afterwards, the publication of the catalogue Hispania Vetus Besides, these pieces are short, their music is quite simple and syllabic and there are (Zapke 2007), and of the colour facsimile of the ‘León Antiphoner’ E-L MS 8 (Fernández none of the lengthy melismas which are so typical of Old Hispanic chant.8 From a de la Cuesta 2011) fostered a renewed interest in Old Hispanic musical manuscripts.3 methodological point of view, when we try to understand Old Hispanic notation by Some scholars focussed on the persistence of some Old Hispanic melodies along the means of its Aquitanian transcription, we must consider a number of issues. Firstly, centuries and across liturgical repertories,4 while others preferred to investigate the Old Hispanic and Aquitanian are two very different styles of notation. In Aquitanian Old Hispanic psalmody5 or the scale of Old Hispanic chant (Huglo 2003). notation a note is expressed as a position on a vertical scale while in Old Hispanic notation a note is expressed as a rising or descending pen-stroke. Compared to other 1. Old Hispanic melodies in diastematic notation notations, we recognise in Aquitanian neumes a loss of musical nuance; indeed, some The Old Hispanic melodies which received most attention are those found also written neumes for special effects such as the virga, bivirga, trivirga, virga strata, the distropha, in Aquitanian notation.6 The manuscript E-Mh Cod. 56 (from San Millán de la Cogolla, tristropha and the pressus are reduced to one or more dots when transcribed into La Rioja) contains sixteen melodies from the Office of the Dead. E-SI ms. 4 (from San Aquitanian notation (Huglo 2011, p. 166). Moreover, Aquitanian notation does not Prudencio de Monte Laturce, La Rioja) contains three antiphons for the Foot-Washing reflect the variety of neume shapes found in Old Hispanic notation. Indeed, groups of two, three, four or more notes written in Old Hispanic notation become simplified when

3. Hornby and Maloy 2013; 2016a; 2016b; 2017; Hornby 2016; Maloy 2014; Deswarte 2013; De Luca 2017a and 2017b; De transcribed into Aquitanian notation because there is no possibility of transcribing Luca and Haines 2017. into this notation the different styles of neumatic connections, which are typical of Old 4. For example, Huglo (1955) discussed some ancient Gallican-Old Hispanic preces which served as model for the preces Hispanic music writing. Furthermore, it may be also relevant to refer to Rodriguez that can be read in some eleventh-century manuscripts from Southern France, in some Missals and Graduals from northern Spain dated soon after the imposition of the Franco-Roman rite, and in the twelfth-century fragment P-G C 623. Fernández Suso’s remarks on the analysis of the chants for the dedication of the altar found de la Cuesta (2013) made an analysis of the dissimilarities between the melodic profiles of the songs in the choir books of Toledo Mozarabic chapel and those of the oldest musical Old Hispanic manuscripts. Ferreira (2013) focussed on the written in more than fifty manuscripts with different notational styles. She pointed out responsory Conclusit vias meas inimicus (Cantus ID 006306) identified as a Gallican survival, itself dependent on an older that every time a melody was transcribed into a different (and later) notational style, Old-Hispanic version. Haggh-Huglo (2013) analysed the chants for the officium of St. Andrew that can be read in the Old Hispanic manuscripts I-VEcap cod. LXXXIX and E-L 8 and in other chant repertories. Gutierrez (2013) discussed the the new version maintained all the corrections made to the elements of the melody melodies of some hymns and preces preserved in tenth- and eleventh-century Old Hispanic manuscripts. These chants can be also found in later Gregorian books and in the sixteenth-century Cisneros cantorales. Ruiz Torres (2015) studied a that were considered irregular in the previous adaptation. Hence, the grouping of newly discovered fragment of Antiphoner copied around 1100 which contains a hitherto unknown office for Saint James melodies according to their notation is not a purely graphical matter but it involves the Apostle. The author discussed the extent to which new Iberian composition for the Franco-Roman rite was open to Old Hispanic melodic structures. some musical meaning as well, because to each notational group corresponded also a 5. On Old Hispanic psalmody see Randel (1969 and 1985); Zapke (1992); Rodriguez Suso (1998); and Ferreira (2006). specific liturgical style and musical taste (Rodriguez Suso 1998, p. 17). 6. On these melodies, mostly of the La Rioja tradition, see: Aubry (1908, pp. 57-75); Prado (1928); Rojo & Prado (1929, pp. 66-82), present twenty-one pieces; alternative transcription of responsory verses in Randel (1969, p. 66); Brockett (1968, p. 109, alternative transcription of ant. Memorare domine). Asensio Palacios (1994) provides the same twenty-one pieces as 7. Facsimile edition of F-Pn lat. 776 in Albarosa (2001). in Rojo and Prado in slightly different transcriptions; Rodriguez Suso (1998); Asensio Palacios (2008, pp. 140-141 and 2009, pp. 19-21). Some of these references are taken from Ferreira (2006, p. 630, note 24). 8. See, for example, the long melisma running along the edge of fol. 203 in E-L MS 8.

20 21 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

2. Old Hispanic neumes: the palaeographers’ point of view notations did.11 Thus, it is wiser to allow the possibility that the same sign may have Huglo and González Barrionuevo studied the relationship between Old Hispanic had different musical meanings in different notational systems, geographically distant notation and other Western European early music writing systems (Huglo 1987 and but chronologically simultaneous. Furthermore, González Barrionuevo’s methodology González Barrionuevo 1995). Huglo focused on the problem of the origins and made cannot clarify whether any shift occurred to the original meaning of the Old Hispanic a broad comparison of the structural elements of notation (ductus, neume shapes, neumes when they were used to accommodate the imported Franco-Roman melodies. ornamental neumes, etc.) among the main Western European early notations, In fact, the transcription of an earlier neumatic tradition into a fixed pitch may include acknowledging the absence of liquescence in Old Hispanic notation. He also discussed some subjective choices made by the scribe, changing or editing the melody. Finally, this the extent to which Old Hispanic culture informed the Carolingian reform of liturgical methodology does not advance our understanding of the melodic grammar of the Old chant, concluding that Old Hispanic notation may even be the precursor of Western Hispanic chant because it investigates de facto the music of the Franco-Roman repertory. European music writing (Huglo 1987, pp. 25-26). On occasions palaeographers devoted their attention to the analysis of a single Old González Barrionuevo is the scholar who made the most systematic and in-depth Hispanic manuscript, such as the case of the León Antiphoner and the San Juan de la analysis of Old Hispanic notation, producing an extensive bibliography on some Peña Antiphoner (E-Zfm M-418, online), studied respectively by González Barrionuevo of its palaeographical features. Being an expert on Gregorian semiology, González (2013) and Zapke (1995). González Barrionuevo (pp. 98-99) noted the elegance and Barrionuevo focused his attention almost exclusively on some hybrid manuscripts calligraphic qualities of the Antiphoner’s notation and its long neumes, usually found produced at the end of the eleventh century (during the transition period between the in chant genres such as the sono, sacrificium and responsory. He pointed out a certain two rites). In those manuscripts the scribes used a set of symbols familiar to them – the tendency toward diastemacy in the manuscript, which seems to be demonstrated by Old Hispanic neumes – to represent the unfamiliar Franco-Roman melodies.9 González the series of consecutive gravis, slanted virga, vertical virga, punctum, tractulus, and Barrionuevo focused especially on the notation of the Antiphoner from Silos GB-Lbl add. uncinus found written within the same piece. He presented also a semiological analysis 30850, but also considered some notational aspects of the Benedictine monastic Breviary of thirteen shapes of the clivis neume found in the León Antiphoner explaining their GB-Lbl add. 30848.10 He compared the Old Hispanic neumes found in these Franco- meaning by means of the comparison of the use of the same (or similar) shapes in other Roman manuscripts with the neumes found in other Western European Franco-Roman Western European early notations (pp. 99-102). Later in the essay, he attempted a further manuscripts written in different notations: I-Lc 601; I-Ra 123; CH-SGs 390-391; F-LA palaeographical and semiological analysis of the Antiphoner’s neumes by means of the 239; GB-WO F. 160, and the Mont-Renaud Antiphoner with Noyon notation – private analysis of two antiphons whose music is also found written in Aquitanian neumes in collection – (González Barrionuevo 1986; 1987a; 1987b; 1990; 1993; 1995; 1997; 2008). the codex Aemilianensis (pp. 109-120). At the beginning of this section the author made explicit the methodology applied here, saying that the analysis of the two melodies is This methodology has the advantage of providing clues to the musical meaning of the Old made in a broad comparative perspective which takes into account tenth- and eleventh- Hispanic neumes using as an interpretative key the neumes adopted in other notations century Old Hispanic manuscripts, some Western European notated manuscripts, the for representing the music of the same Franco-Roman chants. However, this involves a later diastematic versions of the same melodies and also the rhetoric of the text, phrase methodological misconception. Even though Old Hispanic notation shares some basic division, punctuation, and accents (p. 109). From the present author’s point of view, the symbols and general principles with other early notations, we cannot assume that the limitations of using later Franco-Roman manuscripts (Iberian and non-Iberian) to infer shared symbols represented the same musical meaning everywhere, especially if we the meaning of the early-tenth-century neumes of the Antiphoner are essentially two. consider the graphical richness of Old Hispanic notation compared to other early music Firstly, eleventh-century Old Hispanic manuscripts no longer reflected the graphical writings. In fact, Old Hispanic notation has a wider array of signs used to represent variety of the earlier sources, such as the early tenth-century León Antiphoner.12 Secondly, music, and it transmitted more musical nuances and details than other contemporary the pitch-readable chants all survive in manuscripts used in the central-northern parts of Spain (in or near Santo Domingo de Silos, or in the La Rioja region). While the melodies of

9. On these manuscripts see Huglo (1985, pp. 255-256). There still survive some fragments written in Visigothic script and Aquitanian notation which reveal the confusion of Old Hispanic scribes during the transitional period. See, for example, the two-folio fragment of a Franco-Roman Antiphoner now attached to E-Tc Ms. 10.5 (Nelson 2007-2008); about some 11. See, for example, the tiny pen-strokes placed at the top-left corner of neumes in E-L Ms. 8, fol. 57v. fragments from Zamora with similar characteristics, see Nelson (2002). Some more fragments are freely consultable in the Portuguese Early Music Database P-Cua IV 3ªS-Gv. 44 (20) and 44 (21); P-LA Caixa 1, Fragm. 10; P-BRad Pasta dos 12. Regarding the dating of the León Antiphoner, the majority of scholars believe it was written in the middle of the documentos visigóticos, s.s., (fragment now lost); E-TUY Fragm. 3. tenth century. See, inter alia, Millares (1999, p. 69); Zapke (2007, p. 252); Deswarte (2013, p. 69). Earlier datings have been proposed by Díaz y Díaz (2007, p. 99: first third of the tenth century); and Menéndez Pidal (2003, pp. 148-151): ca. 906. Two 10. On GB-Lbl add. 30850 see Fernández de la Cuesta (1976 and 1985a). Palacios (2009, pp. 23-24). GB-Lbl add. 30848 is scholars dated the León antiphonary to 1069: García Villada (1923, p. 198); and Gómez Moreno (1954, pp. 314-317). Zapke said to be written with Aquitanian notation in Zapke (2007, p. 376), however, it is written with Old Hispanic notation with a in a recent essay (2011, p. 94) preferred a more general eleventh century dating. For a reassessment of the dating of the strong tendency towards diastemacy, as it can be seen from the full-page image on p. 377. manuscript to the years 900-905 see De Luca (2017a and 2017b).

22 23 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION these regions are related in broad terms to the more Western Iberian tradition preserved not seem entirely correct to use the Latin terminology to describe the neume shapes in the León Antiphoner, the note-by-note relationship is not always very close (Randel shared with other Western European notational styles, because we cannot assume that 1969). This means that apparent divergences between the pitch-readable notation and those shapes conveyed the same musical meaning in each early notational system. the Antiphoner’s signs very likely reflect different versions of the melodies rather than Latin names for neumes are useful labels given a posteriori to pinpoint musical signs pointing towards a counter-intuitive but melodically identical way of interpreting a and the use of these names for earlier tenth century manuscripts is anachronistic. León Antiphoner sign. Further, since the Antiphoner has many ways of showing many Indeed, the Latin names for musical signs were created by medieval theorists who melodic movements within a single pen-stroke (for example, a combination of a note needed a terminology that would allow them to talk about music in their treatises. followed by a higher note, conventionally known in the scholarship as a pes or podatus), The Latin names are first found in music theory manuscripts written in the eleventh an explanation of the graphic differences found in the León Antiphoner cannot be found and twelfth centuries and, when we compare the lists of neume names found in those through comparison with chants in pitch-readable notation, which uses a much smaller sources, they do not always match (ibidem, pp. 54 ff.). set of signs. In addition, the disadvantage of using external evidence for palaeographical research may cause important information concerning the manuscript being investigated Further, while there are conventional, and widely understood, names in use for neumes to be lost. In fact, when González Barrionuevo discussed the Antiphoner’s notation, he of up to three notes (e.g. pes, clivis, torculus, porrectus, scandicus, climacus), the language stated that a single person wrote the manuscript; moreover, while discussing some becomes more clumsy when four notes are present (for example, torculus resupinus; changes to the original layer of notation, he referred to the ‘notator or corrector’ without scandicus flexus; pes subbipunctis etc.). For the longer neumes, so characteristic of the openly clarifying whether it was the same person (pp. 104-105). As will be discussed León Antiphoner’s notation, this naming of neumes becomes impossible according to below, the Antiphoner shows instead the presence of at least four music scribes and these conventions. There is simply no name for a single pen-stroke with eight, nine, or multiple later hands. even more notes.13 Because of this, and because I am concerned with deciphering the melodic contours of the neumes and their nuances, I have adopted a terminology based Zapke (1995) adopted a similar approach in her study of the San Juan de la Peña Antiphoner on those melodic contours. This approach was already adopted by Hornby and Maloy and, also in this case, the palaeographer used the traditional Latin terminology to refer (2013) but it is here ironed out. In Old Hispanic notation the motion of the melody is the to the neumes (virga, clivis, porrectus etc.). Each neume shape found in the eight-folio key element. The inclination of the pen-stroke tells us whether the melody goes up or fragment is presented along with its semiological explanation, the latter being based on down. With a few exceptions, the ascending and descending movement of the pen-stroke the analysis of the occurrence of the neume shapes with respect to text accentuation. A represents rising and falling pitch (although of course specific intervallic content is not short comparative analysis is presented; in this case it is made using as a comparison communicated). One complicating factor in this notation is that the relationship between the twelfth-century Antiphoner of Santa Cruz de La Serós in Aquitanian notation. the last note of a neume and the first note of the next is not always directly signalled by the inclination and direction of the pen-stroke at the beginning of the second neume. As 3. A different methodological approach a general rule, the first note of any neume has to be interpreted as ‘neutral’ (N) because As an alternative to the methodologies applied in previous research on Old Hispanic we do not know the pitch or its relationship with the previous neume. However, it is notation, I propose the investigation of Old Hispanic neumes when they were used to almost always possible to identify the melodic contour of the following notes of a neume represent Old Hispanic melodies, without constantly referring to other chant repertories using the direction of the pen-strokes: each note is either higher than (H), lower than or notations. This approach emphasizes the value of the study of early notations in (L), or the same as (S) the previous one. Thus, while in the usual modern terminology terms of their own way of functioning and focuses on a single manuscript at a time. a torculus consists of a note followed by a higher and then a lower note, I prefer to identify such a shape as ‘Neutral-High-Low’ . Similarly, modern terminology for a One of the first challenges Old Hispanic notation poses to modern readers is the note followed by two higher notes and then a lower one would be scandicus flexus, but terminology to be adopted to describe its neumes. In Old Hispanic notation we have to I have preferred the label ‘Neutral-High-High-Low’ . A pair of vertical virgae written identify and classify, for the first time, some peculiar palaeographical features whose closely together would be labelled as bilineola; I have labelled them ‘Neutral-Same’ musical meaning is totally obscure. What we know is that those palaeographical . These labels, then, are directly derived from the melodic contours of the neumes, features were used throughout all – or most of – the earlier manuscripts consistently and are nuanced, where necessary in the context, by further information about the and, apparently, for a functional purpose. The well-known Latin names for neumes palaeographical features of the neume in question. (podatus, torculus resupinus etc.) often carry a specific meaning referring to intervals, rhythm or performance (Cardine 1970). For the Old Hispanic repertory however, it does 13. See, for example, the twelve-note neume whose melodic contour is Neutral-High-High-High-Low-High-Low-High- Low-High-Low-Low found in the León Antiphoner on fol. 257v, 16.

24 25 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

The methodology presented here is based on the study of a single manuscript at a The resulting deep knowledge of the constituent palaeographical elements of the time. The first step in approaching the Old Hispanic notation of a manuscript which is Old Hispanic manuscripts analysed can be used to: pinpoint the notational features the object of research is to observe the range of neume shapes used. This initial stage shared among different sources; identify those features which were discarded; and consists of the examination of each neume individually, scrutinising their morphology. identify groups and families of sources which share common features. A major This approach can gradually lead to an understanding of the shapes’ constituent advantage of this methodology is that it permits an understanding of the changes that palaeographical elements, and of how those elements are joined together to create occurred in the Old Hispanic notational system over the centuries during which it the varied neume shapes of the source. The next step is to engage not only with the was used. Furthermore, the combined use of the few securely dated sources and of graphical shapes of the signs, but also with their spatial placement and proximity, the knowledge derived from the palaeographical examination I propose for the Old both vertical and horizontal. It is also important to investigate how the neumes are Hispanic manuscripts will likely result in a new understanding of the geography and combined together, creating short, medium and long melismas on a single syllable or chronology of the manuscripts.15 creating standard melodic patterns sung on more than one syllable at cadence points or repeated within the same or between different chants. Another kind of internal 4. The notation of the León Antiphoner as case-study comparative evidence within the source being investigated is chants written down Being the most substantial notated Old Hispanic manuscript, the León Antiphoner is more than once.14 Obviously, internal comparison provides more information when the best candidate for making a palaeographical analysis of the Old Hispanic notation. the source being investigated is in book format, rather than a fragment. All of these Furthermore, the inner quality of the Antiphoner’s notation promises great potential categories of evidence can be scrutinised to match neume shapes, melodic contour, for palaeographical discoveries. Indeed, the Antiphoner’s notation has a unique and the neumes’ spatial placement and proximity, both vertical and horizontal. It is degree of complexity on account of the wide variety of signs used. The complexity of crucial to check for variations in the precise neume shapes, and then make a careful the León Antiphoner’s notation derives from the number of graphical variations in evaluation of the nature of any varying note forms. For example, variation in specific the basic neume shapes – for example, to notate a three note neume Neutral-High- neume shapes unveiled the presence of several music scribes in the León Antiphoner. Low, more than fifty different shapes are used – and from the presence of graphical elements placed near the notation with the purpose of adding musical information. Finally, another piece of evidence is provided by changes to the manuscript made after All the neume shapes found in the León Antiphoner derive from the combination of its initial copying. These changes may consist of marginal additions to the manuscript, these factors: palaeographical parameters of the pen-stroke (I define a pen-stroke as or they can be a new neuming superimposed on a still-visible original. Some of these a single movement of the pen on the page): orientation, inclination, length; different changes seem to represent a different understanding of the performance nuances and/ kinds of connections between notes within a neume: gapped, angular, curved, and or note groupings which can help us to deduce the meanings of certain signs. looped.

Focusing on a single source at a time, it is possible to become familiar with the The second factor applies only to compound neumes, that is, neumes containing two scribes’ usual pen-strokes. This enables one to identify variants in neume shapes that or more notes, whereas a simple neume can be described in palaeographical terms result from the vagaries of an individual scribe’s hand, rather than those which have only according to the parameters of its pen-stroke. In exploring the ways these factors a specific musical meaning. In addition to this, the identification of different scribes change the morphology of the neumes, I discuss the different kinds of connection first, who participated in neuming a single source is a crucial preliminary step for musical followed by the parameters of the pen-stroke. analysis, because there may be the presence of writing habits peculiar to individual scribes which need to be identified and taken into account while seeking for related I define connection as the way two single pitches are joined together within a neume. neume patterns in cadences, melismas and single notational shapes. This is important There are four different kinds of connections: gapped , angular , curved , due to the great component of orality involved in the transmission of Old Hispanic and looped . Often more than one of these is in evidence at different points within music. In fact, orality was the main means of musical transmission (learning, diffusion a single pen stroke . The consistent use of the connections among the Old Hispanic and preservation) until the mid-eleventh century in all Western Europe (Asensio musical manuscripts tells us that behind each graphical connection there was a Palacios 2009, p. 21). musical meaning, now lost.

15. The few Old Hispanic manuscripts securely dated are: E-L 8; E-SI 3, 4, 5; E-SAu 2668, is a Liber canticorum et horarum, 14. Occasionally chants have been found notated in full two or more times, or (providing more limited information) known as the ‘Breviary of the Queen Sancha’; E-SCu 609 Res.1 known as the ‘Book of Hours of Ferdinand I’; E-Tc 35.4. notated in full once and then signalled a second or more times in the manuscript with a notated incipit. See Randel (1973). Visigothic manuscripts are described in Millares (1999). On the two eleventh-century royal manuscripts see Pick (2013).

26 27 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

The importance of classifying the graphical connection of Old Hispanic notation can be anticlockwise. However, either anticlockwise or clockwise connections can be used better understood if we consider that often we encounter neumes which have the same to join the second and third note of a Neutral-High-Low curved neume ( or ). number of notes and the same melodic contour but different graphical connections (for They differ in the orientation of the final part of the pen-stroke (right to left in the first example, these two-note Neutral-High neumes ). The classification of the neumes example, against the norm, and left to right as usual in the second example). Both according to their connections allows us adequately to describe and differentiate all anticlockwise and clockwise curved connections are used mainly for combinations the neumes, even if they represent the same melodic contour. of notes whose melody rises. A curved connection can be used to join a note to a lower one (as in the Neutral-High-Low neumes shown above), but no neume exists Some neumes are formed of an unbroken pen-stroke; the notes signified within such in the León Antiphoner with a curved connection that includes a fall of two or more neumes are, incontrovertibly, part of the same neume. Other neumes are formed of consecutive notes. Neumes with Low-Low shapes (or even with more consecutive two or more lines of ink on a single syllable, with small gaps between them . This Ls), use angular (staircase) or gapped connections for these falls or both gapped gapped connection is already known in the literature as a way of expressing a vocal and plain angular . articulation (Cardine 1970; González Barrionuevo 1995). I do not necessarily accept that interpretation, but I recognise the phenomenon of the gapped connection within Some neumes with curved connections resemble the shapes found in other notations a single neume. We can only identify separate pen-strokes as part of the same neume as a quilisma (an ornamental neume). There is still lively scholarly debate about under specific and limited circumstances, depending in part on the proximity of the how quilismas were performed. However, there is no direct contemporary theoretical pen-strokes, in part on the way they are combined. commentary on Old Hispanic notation to confirm that these curved connections were ornaments in this culture. Nevertheless, the lack of multiple-note falls with curved An angular connection joins straight lines which can have different lengths and connections might lend weight to the association of the curved connection with the inclinations. In the León Antiphoner we have observed three categories of angular quilisma. connection: plain , , ; v-shaped ; and staircase . These categories seem to be functionally equivalent, with the difference that, if the figure rises or falls for In the León Antiphoner, some neumes have a looped connection at the junction where three or more notes in the same direction, and has an angular connection throughout, the pen-stroke changes direction. When the pen-stroke turns anticlockwise into a then it will use the staircase shapes; if there is a two note rise, it uses a plain shape or loop, the following note is always lower ; when the pen-stroke turns clockwise into a ‘v’shape (which can be combined with further connections to make a longer shape). a loop, the following note is always higher . A wavy punctum is always connected A two-note fall can be given with a plain angular connection . However, if the two- to a higher note by means of looped connection (neutral-Low neumes never have note fall is followed by a further lower note, the whole neume may assume the shape of looped connection between the two pitches). a downward staircase . Plain and staircase connections can be combined within a single neume . The v-shaped connection is very common in the León Antiphoner There is huge variety in the appearance of different pen-strokes within the León although it is used only for upward melodic movement and always connects the first Antiphoner. Orientation (rightwards or leftwards), inclination (from vertical to and second notes of a neume (González Barrionuevo 2008). Although both the plain horizontal), and length are the parameters of the pen-stroke and they can affect the and the v-shaped connection are angular, I do not see them as performatively graphical appearance of the connections produced by the encounter of two sections of equivalent, since considerable scribal effort has gone towards differentiating the pen-stroke or by two separated pen-strokes. If a pen-stroke encompasses multiple between the two. notes, each change of direction results in a new value for each of those parameters. For example, in a compound neume written with a single pen-stroke, the pen-stroke There is great variety in the shape of curved connections within neumes. I have divided changes inclination to signal each new note. Each change of direction results in an angle the curved connections into two categories: in the first, the pen-stroke (assuming it between two lines going in different directions. The width of the angle depends on the is written left to right across the page) curves in an anticlockwise manner ( , ; varying inclination of the pen-stroke as it approaches and leaves the connection. The in the second, the pen-stroke instead curves in a clockwise manner ( , ). The length of the pen-stroke does not affect the connection, but it can change the size and anti­clockwise curved connection is much more common than the clockwise one; proportion of a neume. It is important to remember that variations in the palaeographical Neutral-High curved neumes, for example, only appear with an anticlockwise parameters of the pen-stroke do not change the number of notes involved in a neume connection . In the same way, there is no Neutral-High-Low curved neume with shape. Instead, the parameters appear to reflect different performance possibilities a clockwise connection between the first two notes; this connection is always possibly related to vocal articulation, ornaments, rhythmic nuances etc.

28 29 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

The notation of the León Antiphoner routinely goes from left to right across the page. As already mentioned in the section on clockwise and anticlockwise curved connections, above, in certain circumstances the pen-stroke can go leftwards across the page rather than rightwards ( rather than ). There are other instances where curved connections can go leftwards rather than rightwards on the page (e.g. rather than ) and the same goes for plain angular connections in neumes with the shape Neutral-Low-High-Low Example 6. Fol. 120v, 8 (e.g. rather than ) or Neutral-Low-High-Low-High ( rather than ).

Example 1 Example 2

Example 7. Fol. 157, 2 To have different inclination of the pen-strokes in different notations of similar neumes does not change the kind of connection but it can determine the width of the connection angle. In examples 1 and 2 the neumes have both the same number of Is there a way to work out the importance of the graphical differentiation due to notes and melodic contour (respectively Neutral-Same/High-Low and Neutral-High- length and inclination of the pen-stroke in the notation of the León Antiphoner? Low) but the inclination of the pen-stroke or of some portions of the pen-stroke varies. In some cases changes to original neuming can give us some clues. The changes found in the León Antiphoner can be grouped as it follows: erasure that shorten the original neume shape; erasure of a portion of a chant (which is not replaced with another version); erasure of the original neuming and addition of new neume/s; correction made by overwriting the new neume on the original neuming (which can still be recognized); neume/s added next to the original neuming, which is still Example 3. Fol. 93, 5 Example 4. Fol. 155v, 4 Example 5. Fol. 259v, 13 readable over the text; marginal additions that correct neume/s written over an erasure; dubious corrections: the changes seem to be unrelated to the music content Changes in the length of a pen-stroke (or in portions of the pen-stroke) can change of the page. Let us examine some examples first. Examples 8, 9, 10 and 11 show some the shape of a neume. In example 3 the two consecutive neumes represent the same interventions made to the length of the original neume. In examples 8 and 9 the melodic contour (Neutral-Low) and both have angular connection between the two pen-stroke is lengthened by means of a new section added at the end of the neume. pitches. What differs between the two neumes is the length of the first section of the In examples 10 and 11 a new single-note neume is written above the original one. pen-stroke. Similarly, example 4 (see arrows) shows two three-note Neutral-High-Low neumes that differ for the length of the intermediate pen-stroke. While in examples 3 and 4 it seems possible to recognize an intentionality in differentiating the two neume shapes, the same cannot be said for the lengths of the three consecutive Neutral neumes in example 5 which may be simply on account of the ductus of the hand on the page.

Example 8. Fol. 284v, 13 Example 9. Fol. 269, 16 The inclination and length of the pen-stroke are part of the graphical alphabet of Old Hispanic notation and they were used to differentiate the neumes graphically, as we can observe in the series of one-note Neutral neumes in example 6. Similarly, we observe in example 7 that neumes 1, 2, and 3 all represent the same melodic contour (Neutral-High-Low) but they are graphically differentiated by their intermediate and final pen-strokes which have, respectively, different lengths and different inclination and lengths. Example 10. Fol. 78v, 2 Example 11. Fol. 112v, 2

30 31 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

Examples 12, 13, 14, and 15 instead show some changes to the original neuming that the Antiphoner suffered massive manipulations; later hands intervened heavily on the involve both the length and inclination of the pen-stroke. Finally, in example 16 not only original layer of notation and left no page untouched – some marginal additions and was the length of the original Neutral-High neume changed, but also its connection, erasures were already discussed by Brou (1954), and Randel (1969, p. 11 ff.). which was angular and became curved anticlockwise in the revision. It is possible to identify at least six later hands who made changes to the original neuming: Scribes E, F, G, H, I, and J; but also Scribes A, B, C and D made changes to their neuming. At the current stage of research, the folios of the León Antiphoner that can be securely attributed to the ten identified scribes are:

Example 12. Fol. 80, 4 Example 13. Fol. 84, 14 Example 14. Fol. 96, 3 Scribe A fols. 29-36, 40-48v

Scribe B fols. 36v-40

Scribe C fols. 77– 92v, 94, 232v-234

Scribe D fols. 123, 129-131v, 160v-165, 181-211v, 269, 271v, 273v, 275v

Scribe E fol. 1v

Scribe F fol. 3v Example 15. fol. 96, 10 Example 16. Fol. 269, 2 fols. 5 top, 305v-306 (chants added over erasure), 64 (melisma on the side), Scribe G 72v, 3 (melisma), 74v (sacrificium over erasure), 132 (melisma), 189 (melisma), All these changes are extremely important for our modern understanding of the 251 (sono) parameters of the pen-stroke in the León Antiphoner’s notation as they demonstrate Scribe H fol. 5 (bottom) that scribes used these parameters to convey purposefully some musical meaning to Scribe I fols. 108v (melisma), 195 (melisma) the reader.16 Besides, they demonstrate that the scribes put a great deal of effort trying to capture as much information as possible in the notation in order to guide the reader Scribe J fols. 191v, 207v, 210v, 224 (melismas on the side on all four pages). in useful directions. Finally, I have already acknowledged that the variations found in the basic neume shapes in the León Antiphoner’s notation are greater than what is customary in other western notations. Now, it can finally be clarified that this graphical Although the original notators (Scribes A, B, C and D) share very similar handwritings variety is due to the wide range of application of the parameters of the pen-stroke. and were very disciplined, there are still some graphical details that permit us to identify and distinguish them. One of the differences between three of the Antiphoner’s original The methodology discussed here, applied to the study of the León Antiphoner, led to music scribes is the way they wrote the melodic movement represented by an angular the discovery of the presence of several scribes in the León Antiphoner (De Luca, n.d.). downward staircase. Scribe A usually wrote a deep curve at the top of the staircase The overwhelming quantity of graphic details in the Antiphoner’s notation is probably , and only occasionally wrote even steps , which are a peculiarity of Scribe D, who the reason why previous scholars did not identify the presence of multiple music always put special care into writing even steps and is very consistent in his handwriting. scribes. In fact, if we juxtapose two images rich in graphic details, it can be a challenge finding any idiosyncrasies within them. A team of highly skilled scribes was recruited On the contrary, Scribe B turned the angular steps into curves and never wrote an to write the León Antiphoner. The main body of the Antiphoner was the product of a angular step. Unlike Scribe A, Scribe D’s folios never contain neumes with a deep small group of people, presumably within a relatively short space of time. Scribes A, curve at the top of the staircase. Scribe D has a very elegant script, a steady hand, B, C, and D were the original notators; however, the analysis of the original layer of and perfect control of the space and the size of the neumes. His neumes are generally notation of the Antiphoner is still in progress and it cannot be excluded that there will small and carefully written. Scribe B is the least skilled of the Antiphoner’s scribes. He be more scribes who participated to the original neuming. The original neuming of barely wrote three folios and it appears that he was soon fired, probably because of his lack of experience. Indeed, Scribe B’s (few) folios show many signs of erasures and 16. Further examples of changes to the length and inclination of the original neuming inter alia can be found on fols. 121, 13 (ad te) 127, 16 (factus es) 173, 15 (consurgit) 177, 3 (alleluia) 180, 3 (quoniam); 261v, 10 (deus) 272, 12 (elegite). corrections. Scribe B sometimes made those corrections himself; on occasions they

32 33 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION were made by Scribe A and at other times by later scribes. Scribe B also had trouble Conclusions writing this neume . He always wrote the vertical stem first and then added a circle As with other early notations, Old Hispanic notation entails a large component of to the right. The other scribes instead wrote the same neume with a single pen-stroke orality. Previous scholars circumnavigated the obstacles posed by the non-pitch (on occasions, also Scribe D wrote this neume with two separate pen-strokes). Scribe readable Old Hispanic notation by focusing their attention on the Old Hispanic C has a peculiar way of lengthening some neumes that end with a higher note and melodies found in diastematic notation and using as an interpretative key the neumes he often oversteps the space devoted to notation, even crossing the text line above. found in the eleventh-century Franco-Roman manuscripts written with Old Hispanic Scribe C’s way of crossing the vertical notation space is visually striking and, to a notation and in some other coeval manuscripts written with other notational styles. In certain extent, even disturbing, as it clashes with the beauty and elegance of the rest this chapter I discussed the strengths and weaknesses of the methodologies previously of the Antiphoner (for example, fol. 77v written by Scribe C). The close similarities applied to the study of Old Hispanic notation. Subsequently, I proposed an alternative between the scribes’ handwritings hint that they may have been trained in the same methodological approach based on the study of Old Hispanic neumes when they scriptorium. Scribe B was probably the youngest, while Scribe A was possibly the oldest. were used to represent Old Hispanic melodies (without constantly referring to other Perhaps at some point poor Scribe A passed away because after he started the copy of chant repertories or notations). Pinning down the palaeographical features of each the Antiphoner and wrote about sixty pages (with a short hiatus where we encounter Old Hispanic manuscript and identifying how the neumes were combined together Scribe B) we never see his idiosyncratic neumes again. Instead we encounter for the to create repetitive neumatic patterns at cadences and melismas will ultimately lead first time Scribe C, who wrote about five quires before Scribe D took over. Scribe D us toward a better understand the melodic grammar of the Old Hispanic chant itself. wrote the vast majority of the Antiphoner but he benefitted from the help of Scribe C and other – possibly one or two – scribes to complete the task. When, for example, later The new methodology discussed here has been applied to the study of the León in the Antiphoner we encounter Scribe C again (e.g. fols. 232v-234), it can be seen that Antiphoner. This palaeographical investigation allows us for the first time to discuss in the extra length of his neumes is less pronounced. One can still recognize his habit of great detail some characteristics of its notation and its scribes. The level of detail and care crossing the text line above, but it happens less often. One also receives the impression put in the León Antiphoner’s notation was shown to be exceptional. While the precise that the size of the neumes is under much greater control. melodic meaning of many of the palaeographical elements of the León Antiphoner remains inaccessible to us, the graphical richness of the Antiphoner’s notation hints The astonishing similarity between duplicated chants written either by different or that the scribes were purposefully using connections and the parameters of the pen- the same original notators demonstrate that they were carefully copying from a model stroke to help the reader recall the melody to be sung. Furthermore, the presence (or from models), and that the scribes of the León Antiphoner were very disciplined. of multiple music scribes (both original notators and later hands) undermines the As a rule, musical differences between cognate versions (written either by the same previous assumption of considering the Antiphoner a whole and homogenous witness or different scribes) do not alter the number of notes, or the melodic contour, which of Old Hispanic chant and demonstrates instead the importance of peeling away all usually match. The differences are limited to the choice of neumatic connection and to the chronological layers of musical changes and reconstructing their specificities one the presence of the ornamental neume wavy punctum. by one.18

Among the later hands, Scribe G deserves a special mention. He can be considered as the main corrector of the León Antiphoner because he is the later scribe who made the largest number of alterations to the notation. Scribe G has a very peculiar writing style – he tended to write very angular lines, thick pen-strokes, and rigid neume shapes. Peculiar also was his way of writing the Neutral-Same curved-gapped neume (the rounded and symmetric pen-strokes we would expect are turned into sharp angles by Scribe G). Scribe G is not only the scribe who intervened most in the original notation of the León Antiphoner but further comparative palaeographical analysis demonstrated that he was also the music scribe of the eleventh-century Liber canticorum et horarum of Queen Sancha.17 18. As said (p. 19, note), the preliminary research for this article has received funding from the European Research Council under the European Union’s Seventh Framework Programme (FP7/2007-2013)/ERC grant agreement no. 313133. Further research was funded by the Centro de Estudos de Sociologia e Estética Musical (Cesem - Nova Fcsh) as part of the project UID/EAT/00693/2013 (funded by the FCT/MCTS) and by the Social Sciences and Humanities Research Council of 17. On E-SAu 2668 see above, note 15. The analysis of Scribe G in E-L 8 and E-SAu 2688 can be found in De Luca (n.d.). Canada (SSHRC).

34 35 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

Bibliography Fernández de la Cuesta Ismael (1976), Notas paleograficas al antifonario silence del Museo Britanico (Mss Add. 30,850) in Pérez de Úrbel Justo, Homenaje a fray Justo Pérez de Úrbel, Burgos: University of Albarosa Nino - Rumphorst Heinrich - Turco Alberto eds. (2001), Il cod. Paris Bibliothèque Nationale de Burgos-Abadia de Silos (Studia Silensia 3), pp. 233-256. France lat. 776 : sec. XI. : graduale di Gaillac, Padova: La Linea [Facsimile]. – ed. (1985a), Antiphonale silense: British Library Ms. Add. 30.850, Madrid: Sociedad Española de Álvarez Martínez Rosario - Ismael Fernández de la Cuesta - Ana Llorens Martín eds. (2013), El canto Musicología [Facsimile]. mozárabe y su entorno: Estudios sobre la música de la liturgia viejo hispánica, Madrid: Sociedad Española de Musicología. – (1985b), ‘La irrupción del canto gregoriano en España. Bases para un replanteamiento’, Revista de Musicologia 8/2, pp. 239-248. Asensio Palacios Juan Carlos (1994), El canto de la antigua liturgia hispana in Chant Mozarabe, CD, Colombes (France): Jade, pp. 2-28. – ed. (2011), Liber Antiphonarium De Toto Anni Circulo a Festivitate Sancti Aciscli Usque Ad Finem: Librum Ikilani Abbati: Cathedralis Ecclesiae Legionensis in Hispania Codex Signatus Nr. Viii, Madrid: – (2008), Liturgia y música en la Hispania de la alta edad media: el canto visigótico, hispánico o mozárabe Instituto Nacional de las Artes Escenicas y de la Musica - León: Cabildo de la Santa Iglesia Catedral de in Pedro Calahorra Martíne - Luis Prensa Villegas eds., XI Jornadas de Canto Gregoriano: De la León [Facsimile]. monodia a la polifonía. XII Jornadas de Canto Gregoriano: Pervivencia de la tradición monástica en el pueblo, Zaragoza: Institución Fernando el Católico, pp. 135-156. – (2013), A la búsqueda de las melodías del canto viejo-hispánico: los libros corales mozárabes de 1500 in Álvarez Martínez et al. 2013, pp. 593-646. – (2009), Los manuscritos visigóticos con notación musical de la cantilación al melisma in Pedro Manuel Cátedra García - Eva Belén Carro Carbajal - Javier Durán Barceló eds., Los códices literarios de Ferreira Manuel Pedro (2006), Notation and Psalmody: A Southwestern Connection? in László Dobszay ed., la Edad Media: interpretación, historia, técnicas y catalogación, San Millán de la Cogolla: Instituto Cantus Planus: Papers Read at the 12th Meeting of the IMS Study Group, Lillafüred/Hungary, 2004. Biblioteca Hispánica del Cilengua, pp. 17-29. Aug. 23-28, Budapest: Hungarian Academy of Sciences, pp. 621-639.

Aubry Pierre (1908), Iter Hispanicum. Notices et extraits de manuscrits de musique ancienne conserves dans – (2013), Pervivencia y reconstrucción de un responsorio del Antifonario de León in Álvarez Martínez et les bibliothèques d’Espagne, Paris: Geuthner (https://archive.org/details/iterhispanicumno00aubr). al. 2013, pp. 489-502.

– (2016), ‘Emulation and Hybridisation in Iberia: A Medieval Background’ in Manuel Pedro Ferreira Brockett Clyde Waring (1968), Antiphons, Responsories and Other Chants of the Mozarabic Rite, New York: The Institute of Mediaeval Music (Musicological Studies 15). ed., Musical Exchanges 1100-1650: Iberian connections, Kassel: Edition Reichenberger, pp. 3-17.

García Villada Zacarias (1923), Paleografia Española, 2 vols., Madrid: Publicaciones de la Revista de Brou Louis (1954), ‘Le Joyau des antiphonaires latins’, Archivos Leoneses 8, pp. 7-114. filología española, 1: Texto. – (1955), ‘«Notes de paléographie mozarabe». Manuscrits de Tolède en notation du nord’, Anuario Gómez Moreno María Elena (1954), ‘Las miniaturas del Antifonario de la Catedral de León’, Archivos Musical 10, pp. 23-44. leoneses 8, pp. 300-317.

Cardine, Dom Eugène (1970), Sémiologie grégorienne, Solesmes: Editions de Solesmes. González Barrionuevo Herminio (1986), ‘Presencia de signos adicionales de tipo melódico en la notación «mozárabe» del norte de España’, Revista de musicología 9/1, pp. 11-28. Casares Rodicio Emilio - Fernández de la Cuesta Ismael - Lopez-Calo Jose eds. (1987), España en la Música de Occidente. Actas del congreso internacional celebrado en Salamanca: 29 de octubre - 5 de – (1987a), El «pes initio debilis» en las principales familias neumáticas gregorianas in José López-Calo noviembre de 1985, Año Europeo de la Música, 2 vols., Madrid: Ministero de Cultura, Instituto Nacional et al. 1987, 1, pp. 69-74. de las Artes Escénicas y de la Música. – (1987b), Una grafía particular del porrectus en la notación «mozárabe» de tipo vertical in Emilio De Luca Elsa (2017a), ‘Musical Cryptography and the Early History of the «León Antiphoner»’, Early Music Casares Rodicio et al. 1987, 1, pp. 75-90. History 36, pp. 105-58. (1990), ‘Dos grafías especiales del «scandicus» en la notación «mozárabe» del norte de España’, Revista – (2017b), ‘Royal Misattribution: Monograms in the León Antiphoner’, Journal of Medieval Iberian de Musicología 13/1, pp. 11-79. Studies 9/1, pp. 25-51 (http://www.tandfonline.com/doi/full/10.1080/17546559.2015.1101521). – (1993), ‘Dos «Scandicus Subbipunctis» particulares en la notación «mozárabe» de tipo vertical’, – The Neumes of the León Antiphoner: Written and Oral Transmission in Old Hispanic Chant (n.d.), Anuario musical: Revista de musicología del CSIC 48, pp. 47-62. unpublished manuscript. – (1995), ‘Relación entre la notación «mozárabe» de tipo vertical y otras escrituras neumáticas’, Studi De Luca Elsa - Haines John (2017), Medieval Music Notes as Cryptography in Katherine Ellison - Gregoriani 11, pp. 5-112. Susan Kim, eds., A Material History of Medieval and Early Modern Ciphers: Cryptography and the History of Literacy. New York-London, Routledge (Material Readings in Early Modern Culture), – (1997), ‘Algunos rasgos fundamentales de la notación «Mozárabe» del Norte’, Revista de musicología pp. 30-47. 20, pp. 37-49.

Deswarte Thomas (2013), ‘Polygraphisme et mixite graphique. Note sur les additions d’Arias (1060-1070) – (2008), ‘El pes corto en «uve» de la notación «mozárabe» de tipo vertical’, Inter-American Music Review dans l’Antiphonaire de León’, Territorio, Sociedad y Poder 8, pp. 67-84. 18, pp. 17-72.

Díaz y Díaz Manuel Cecilio (2007), Some Incidental Notes on Music Manuscripts in Zapke 2007, pp. 93-111. – (2013), La notación del Antifonario de León in Rosario Álvarez Martínez et al. 2913, pp. 95-120.

36 37 CANTUS PLANUS 2014 DE LUCA — A METHODOLOGY FOR STUDYING OLD HISPANIC NOTATION

Gonzálvez Rámon (1985), The Persistance of the Mozarabic Liturgy in Toledo after A.D. 1080 in Bernard F. Nelson Kathleen E. (1996), Medieval Liturgical Music of Zamora, Ottawa: Institute of Mediaeval Music. Reilly ed., Santiago, St. Denis, and St. Peter: The Reception of the Roman Liturgy in León–Castile in 1080, New York: Fordham University Press, pp. 157-185. – (2002), Two Twelfth-Century Fragments in Zamora: Representatives of a Period of Transition in David Crawford and G. Grayson Wagstaff eds., Encomium Musicae: Essays in Honor of Robert J. Snow, Gutierrez Carmen Julia (2013), Melodías del canto hispánico en el repertorio litúrgico poético de la Edad Hillsdale, NY: Pendragon Press, pp. 161-174. Media y el Renacimiento in Álvarez Martínez et al. 2013, pp. 547-575. – (2007-2008), ‘Observations on an Early Twelfth-Century Antiphoner Fragment at Toledo’, Inter- Haggh-Huglo Barbara (2013), The Chant for St. Andrew in the León Antiphoner: A Crux of Early Mediaeval American Music Review 17-18: Robert Stevenson and Emilio Rey Garcia eds., Concordis Modulationis Chant Transmission? in Álvarez Martínez et al. 2013, pp. 413-454. Ordo Ismael Fernández de la Cuesta In Honorem, pp. 17-24.

Henriet Patrick (2017), ‘Cluny and Spain before Alfonso vi: remarks and propositions’, Journal of Medieval Pick Lucy K. (2013), ‘Rethinking Cluny in Spain’, Journal of Medieval Iberian Studies 5, pp. 1–17. Iberian Studies 9/2, pp. 1-14. Prado Germán (1928), ‘Mozarabic Melodies’, Speculum 3, pp. 218-238. Hornby Emma (2016), ‘Musical Values and Practice in Old Hispanic Chant’, Journal of the American Musicological Society 69/3, pp. 595-650. – (1934), Cantus Lamentationum pro ultimo Triduo Hebdomadae Majoris, Tournai: Desclée & Socii.

Hornby Emma - Maloy Rebecca (2013), Music and Meaning in Old Hispanic Lenten Chants: Psalmi, Threni Randel Don Michael (1969) The Responsorial Psalm Tones, Princeton, NJ: Princeton University Press and the Easter Vigil Canticles, Woodbridge: Boydell (Studies in Medieval and Renaissance Music 13). – (1973), Index to the Chant of the Mozarabic Rite, Princeton, NJ: Princeton University Press. – (2016a), ‘Biblical commentary in the Old Hispanic liturgy: a Passiontide case study’, Early Music 44/3, pp. 383–394. – (1985) ‘El antiguo rito hispanico y la salmodia primitiva en Occidente’, Revista de Musicologia 8/2, pp. 229-238. – (2016b), ‘Melodic Dialects in Old Hispanic chant’, Plainsong and Medieval Music 25/1, pp. 37-72. Reilly Bernard F. ed. (1985), Santiago, St. Denis, and St. Peter: The Reception of the Roman Liturgy in León- – (2017), ‘Fixity, Flexibility, and Compositional Process in Old Hispanic chant’, Music & Letters 97/4, Castile in 1080, New York: Fordham University Press. pp. 547-574. Rodríguez Suso Carmen (1992), ‘El manuscrito 9 del monasterio de Silos y algunos problemas relativos a Huglo Michel (1955), ‘Les Preces des graduels aquitains empruntées à la liturgie hispanique’, Hispania la adopción de la liturgia romana en la península Ibérica’, Revista de Musicología 15, 2/3: La Música Sacra 8/16, pp. 361-383. en la Abadía de Silos: Trabajos de I Simposio de Musicología Religiosa, pp. 473-510.

– (1985), ‘La penetration des manuscrits aquitains en Espagne’, Revista de Musícologia 8/2, pp. 249-256. – (1998), ‘L`évolution modale dans les antiennes de l’ordo wisigothique pour la consécration de l’autel’, Études grégoriennes 26, pp. 173-204. – (1987), La Notation wisigothique est-ele plus ancienne que les autres notations européennes? in Casares Rodicio et al. 1987, 1, pp. 19-26. – (2004), Les chants pour la Dédicace des Églises dans les anciennes liturgias de la Septimanie: leur contexte liturgique et leur transmission musicale in Christian-Jacques Démollière ed., L’Art du Chantre – (2003), The Diagrams Interpolated into the Musica Isidori and the Scale of Old Hispanic Chant in Carolingien. Découvrir l’esthétique première du chant grégorien, Metz: Éditions Sepenoise, pp. 91-101. James Haar - Sean Gallagher - John Nádas - Timothy Striplin eds., Western Plainchant in the First Millennium: Studies in the Medieval Liturgy and its Music, Aldershot: Ashgate, pp. 243-259. Rojo Casiano - Prado Germán (1929), El Canto Mozarabe e studio histórico-crítico de su antigüedad y estado actual, Barcelona: Diputación Provincial. – (2011), The earliest developments in square notation: twelfth-century Aquitaine in John Haines ed., The calligraphy of medieval music, Turnhout: Brepols (Musicalia Medii Aevi 1), pp. 163-171. Rubio Sadia Juan Pablo (2011), La recepción del rito francorromano en Castilla (ss. XI-XII). Las tradiciones litúrgicas locales a través del Responsorial del Proprium de Tempore, Città del Vaticano: Libreria Levy Kenneth (1984), ‘Toledo, Rome, and the legacy of Gaule’, Early Music History 4, pp. 49-99. Editrice Vaticana.

– (1987), Old-Hispanic Chant in its European Context in Casares Rodicio et al. 1987, 1, pp. 3-14. Ruiz Teofilo F. (1985), Burgos and the Council of 1080 in Bernard F. Reilly ed., Santiago, St. Denis, and St. Peter: The Reception of the Roman Liturgy in León–Castile in 1080, New York: Fordham University Maloy Rebecca (2014), ‘Old Hispanic Chant and the Early History of Plainsong’, Journal of the American Press, pp. 121-130. Musicological Society 67/1, pp. 1-76. Ruiz Torres Santiago (2015), ‘¿Vestigios del corpus viejo-hispánico en la composición ibérica de canto Menéndez Pidal Gonzalo (2003), Varia medievalia ii. Estudios alfonsies - Cifras y caminos, Madrid: Real llano? El oficio pre-Calixtino de Santiago apostol’, Revista de musicología 38, pp. 395-417. Academia de la Historia. Vones Ludwig (2007), The Substitution of the Hispanic Liturgy by the Roman Rite in the Kingdoms of the Millares Carlo Augustin - Manuel Cecilio Díaz y Díaz - Anscari Manuel Mundó Marcet - José Manuel Iberian Peninsula in Zapke 1992, pp. 43-59. Ruiz Asencio - Blas Casado Quintanilla - Enrique Lecuona Ribot eds. (1999), Corpus de códices visigóticos, Las Palmas de Gran Canaria: Universidad de Educación a Distancia, Centro Asociado de Walker Rose (1998), Views of Transition Liturgy and Illumination in medieval Spain, London: The British Las Palmas de Gran Canaria. Library - Toronto: University of Toronto Press.

38 39 CANTUS PLANUS 2014

Zapke Susana (1992), ‘Estructura melódica de la salmodia responsorial del rito hispánico, ejemplificada en el antifonario de San Juan de la Peña’, Nassarre 8, pp. 155-184.

– (1995), El antifonario de San Juan de la Peña (siglos X-XI): estudio litúrgico musical del rito hispánico, Zaragoza: Institución Fernando el Católico.

– ed. (2007), Hispania Vetus: Musical-liturgical Manuscripts from Visigothic Origins to the Franco- Roman Transition (9th-12th centuries), Bilbao: Fundación Bbva.

_ (2011), Dating Neumes According to their Morphology: The Corpus of Toledo in John Haines ed., The Calligraphy of Medieval Music, Turnhout: Brepols (Musicalia Medii Aevi 1), pp. 91-99. II Manuscripts, fragments, and sources

Sitography

Biblioteca Digital Real Academia de la Historia http://bibliotecadigital.rah.es/dgbrah/i18n/catalogo_imagenes/grupo.cmd?path=1000098 [Manuscript E-Mh Cod. 56]

Biblioteca Virtual del Patrimonio Bibliográfico http://bvpb.mcu.es/es/catalogo_imagenes/grupo.cmd?path=26408&presentacion=pagina®istrar download=0&posicion=411 [Manuscript E-L MS 8]

Gallica, Bibliothèque National de France http://gallica.bnf.fr/ark:/12148/btv1b84546727 [Manuscript F-Pn lat. 776]

Portuguese Early Music Database http://pemdatabase.eu/source/35895 [Manuscript E-TUY Fragmento 3] http://pemdatabase.eu/source/12692 [Manuscript P-BRad Pasta dos documentos visigóticos, s.n. (fragment now lost)] http://pemdatabase.eu/source/266 [Manuscript P-Cua IV 3ªS-Gv. 44 (20)] http://pemdatabase.eu/source/155 [Manuscript P-Cua IV 3ªS-Gv. 44 (21)] http://pemdatabase.eu/source/205 [Manuscript P-LA Caixa 1, Fragmento 10]

Universidad de Salamanca https://gredos.usal.es/jspui/handle/10366/55563 [Manuscript E-SAu 2668]

Universidad de Zaragoza https://zaguan.unizar.es/deepZoom2/antifonario/index.html#pagina1 [E-Zfm M-418]

40 41 David Catalunya

The ‘codification’ of new Latin song in early twelfth century Aquitania: codicological insights into F-Pn fonds latin 1139

As noted in previous papers from this session, undertaking to produce a ‘source edition’ presupposes that the editor has an intimate familiarity with the original manuscript itself. One of our main concerns in the Corpus monodicum edition of twelfth-century Latin song (versus) was to determine the nature of the sources that transmit the repertoire. As a taste of what will be the introductory study of our new edition, I am presenting here a brief overview of the codicology of the earliest of the Aquitanian manuscripts thattransmit the repertoire, F-Pn lat. 1139.1

Let me begin by putting our manuscript in a historiographical perspective. Along with F-Pn lat. 1139, three other manuscripts transmit the Aquitanian repertory of twelfth- century versus both monophonic and polyphonic: F-Pn lat. 3719, F-Pn lat. 3549, and GB-Lbl Add. 36881. Together, the four manuscripts were traditionally regarded as the main sources of a ‘school of polyphony’ presumed to have existed in Saint Martial of Limoges in the twelfth-century. However, analysis for her 1970 doctoral dissertation led Sarah Fuller to conclude that these four codices consist of no fewer than nine discrete manuscripts, or versaria. She argued that the sources are in fact composite codices or miscellaneous collections of manuscripts of different origins. Divergences among the versaria, reasoned Fuller, point to their origin in a less unified group of sources, thus calling into question the very existence of a specific ‘Saint Martial School’. The versaria, she posited, may have been manuscripts made by individuals for their own personal use.

Fuller’s dissertation has greatly influenced the study of Aquitanian polyphony ad played a major role in shaping our current historical view of the repertory. Since its publication, many scholars have unquestioningly embraced her versaria theory to guide their own research. For instance, James Grier’s description (1985, p. 33; 1988, p. 252) of the source panorama in his 1985 study emphasizes that

each of the [nine] discrete manuscripts is distinct and appears to be a personal book, copied for and perhaps by a specific singer for his own use […] Perhaps the best term to describe these independent manuscripts is libellus.

1. This paper is an extremely condensed version of a larger work that forms part of the introductory study to the Corpus monodicum edition of the twelfth-century repertory of Aquitanian versus (Catalunya and Voigt, forthcoming), to which the reader is directed for more extensive discussion and details.

43 CANTUS PLANUS 2014 CATALUNYA — THE ‘CODIFICATION’ OF NEW LATIN SONG IN EARLY TWELFTH CENTURY AQUITANIA

Hendrik van der Werf perpetuated this view in his edition of 1993, and later scholars presents damage at the edges of the folios, the edges of the α section are intact. On the reinforced the notion that the Aquitanian repertory of versus circulated through other hand, while the folds of the β gatherings are intact, those of the α gatherings are a process of ‘gathering manuscripts’ or of small libelli that were assumed to be the consistently and homogeneously damaged. It is clear then that the damage to the α personal collections of individual singers (Werf 1993). gatherings could not have been caused by the current binding or by that of 1265, but by an earlier and defective binding that put too much stress on the folds of the gatherings. The libellus-theory undoubtedly helps to explain in part the origin of some of the gatherings, or quires, that are bound together today as part of the composite manuscript F-Pn lat. 3719, or even of GB-Lbl Add. 36881. Nevertheless, in my view, the subdivisions of the earlier corpus of F-Pn lat. 1139 were over-interpreted in order to force the manuscript to fit within the narrative of the libellus-theory model of transmission. Fuller (and all scholars following her) treated the oldest corpus of the codex as three «independent» manuscripts: 1139a, 1139b, and a proser. During a research trip to the Bibliothèque Nationale, I had the opportunity to re-examine the original manuscript; here I present some of the results of my own ‘autopsy’ of the manuscript, which tell a somewhat different story.

In its current state, the composite codex F-Pn lat. 1139 consists of thirty gatherings, which can be divided into three main sections of different origins and varying dates. I will call these sections α, β, and γ. Section α consists of eleven gatherings (fols. 32-118) and is the earliest corpus of the manuscript, assumed to have been written around 1100. This section corresponds to the three manuscripts that Fuller and Grier referred to as 1139a, 1139b and an «independent proser».2 It contains a variety of genres, including liturgical plays, proses, a kyriale, and the collection of versus for which it is typically known. Section β (fols. 9-31 and 119-228) dates from the early thirteenth century and contains a new collection of proses, Marian responsories, antiphons, and several documents from the abbey of Saint Martial written on residual folios. Section Figure 1. Codex F-Pn lat. 1139, detail of the binding. γ (fols. 1-8) consists of later additions to the β corpus and contains other proses added Photo: David Catalunya, with permission by the Bibliothèque Nationale de France by several hands that can be dated to around the middle of the thirteenth century. The last gathering of the codex (fols. 229-235) contains a booklist from the library of the abbey of Saint Martial. The parchment is absolutely consistent throughout all eleven of the α gatherings. While the quality of the skins (likely goat) was rather poor, the fabrication and the treatment The current configuration of the composite codex dates from 1265, when it was bound of the parchment were extraordinarily refined.3 This suggests that considerable effort at the abbey of Saint Martial, likely by the librarian Helias de Brolio. The α gatherings was made to improve and optimise the material quality of the book. The preparation are of much narrower physical dimensions than the others in the composite codex. of the gatherings (pricking and ruling) is equally consistent throughout the entire α As shown in figure 1, the codex was restored in the late twentieth century at the corpus; all of the gatherings were prepared using the same tools and techniques at a 4 Bibliothèque Nationale, and a sort of frame was added to each gathering in the α single scriptorium. section to make them fit into the binding. This restoration did not mask the fact that the The first six gatherings were written mainly by one hand. This scribe A interrupted his different sections of the codex show different patterns of damage. While the β section work at the beginning of gathering 7, where the collection of proses starts. A second scribe, scribe B, completed the proses that scribe A left incomplete, and continued 2. What Fuller calls 1139a is the second gathering of the earlier corpus of F-Pn lat. 1139. Although this gathering (fols. 40-47) was mainly copied by ‘scribe A’, it presents a different pattern of concordances, displays the versus in a dissimilar mise en page, and lacks a quire signature. This certainly suggests that the gathering was inserted at a later stage of the 3. The thickness of the parchment measures just 0.2 millimetres on average and is still surprisingly supple for its age. compilation process, likely because a new exemplar had become available. Nevertheless, the content of this supplementary gathering was completed before the entire codex was rubricated (see below). 4. Scribe B, however, preferred a different number of lines per page.

44 45 CANTUS PLANUS 2014 CATALUNYA — THE ‘CODIFICATION’ OF NEW LATIN SONG IN EARLY TWELFTH CENTURY AQUITANIA the compilation of the proser and the kyriale (see table 1). We will never know why In the case of F-Pn lat. 1139, I believe that this proposed scenario of the compilation of a scribe A stopped where he did; perhaps he moved to another project, or he fell ill. codex might be related to the different scribal behaviours encountered throughout the However, the abrupt replacement of one scribe by another is very telling, for it reveals manuscript.6 This is not a case of a singer making copies for personal use, but rather that they were not working on «discrete manuscripts», but rather on a single, well- of a professional scribe who has been entrusted with a list of pieces to copy. The scribe planned project that needed to be brought to completion. The scribes must not have may have been more familiar with some pieces than with others. Internal evidence been working for themselves, but were likely commissioned for the creation of an suggests that he may have been relying in turn on different types of written exemplars institutional codex. (of different degrees of readability) and on his own memory or personal experience for pieces with which he was more familiar. In any case, it is clear that the codex is not Table 1 a mere copy of another, similar codex. Thus, F-Pn lat. 1139 may represent one of the Shift of scribes at the beginning of the proser in F-Pn lat. 1139 first attempts to systematize the new repertory of versus within a larger manuscript.

Folio Incipit Text Scribe Music Scribe

80-80v Nato canunt A A / B (Scribe B completes the end)

80v -81v Letabundus A A / -- (without music from 81v onwards)

81v-82 celebranda A B

83-84 Epiphaniam B B

Once the scribes had completed their work (primarily scribes A and B, but other secondary hands as well), a single rubricator provided the rubrics for all eleven gatherings of α. Scribes A and B left the required space for the rubrics, and included marginal indications of the texts to be added. The rubricator’s task was to provide the necessary rubrics as a locating or indexing device, which, in turn, contributed to unifying the visual aspect of the entire codex. The rubricator took on the additional task of proofreading the copy, providing suitable clarifications and even corrections.5

Bibliography To conclude, physical evidence gleaned from close observation of the manuscripts, particularly evidence concerning the compilation process, suggests that the earliest Catalunya David - Voigt Konstantin eds. (forthcoming), Lieder aus Quellen des 12. Jahrhunderts, Basel: Schwabe (Corpus monodicum 4). corpus of F-Pn lat. 1139 (fols. 32-118) is not a collection of three independent libelli that were collated into a single miscellaneous codex. Instead, the manuscript seems Grier James (1985), Transmission in the Aquitanian Versaria of the Eleventh and Twelfth Centuries, Ph. D. dissertation, University of Toronto. to have been planned and executed as a proper and unified codex that was bound and had existed as such for a century prior to its later inclusion within a miscellaneous – (1988), ‘The Stemma of the Aquitanian Versaria’, Journal of the American Musicological Society 41/2, collection. The compilation of a well-planned codex requiring a process of teamwork pp. 250-288. presents quite a different scenario from that envisaged by Fuller, Grier, van der Werf – (1990), ‘Some Codicological Observations on the Aquitanian Versaria’, Musica Disciplina 44, pp. 5-56. and many other scholars for F-Pn lat. 1139 (fols. 32-118). My re-examination of the Fuller Sarah (1970), Aquitanian Polyphony of the Eleventh and Twelfth Centuries, Ph. D. dissertation, three other Saint Martial manuscripts reveals a greater diversity in the type of sources University of California. that transmit the repertory: not only do they include small personal collections, but also sources that manifest the 'codification' of the repertory within larger manuscripts. van der Werf Hendrik (1993), The Oldest Extant Part Music And The Origin of Western Polyphony, 2 vols., Rochester, NY, edited by the author.

5. The rubricator even added corrections and clarifications to the musical notation (see, for example, fols. 46r and 55v). The role of the rubricator as a proofreader and corrector is also observed in F-Pn lat. 3549 and some Aquitanian tropers. 6. This variety of scribal behaviours is further explored in our Corpus monodicum edition.

46 47 Marco Gozzi

Manuscripts in Cortona: fragments and liturgical books in the Archivio storico diocesano

In this paper I wish to present an important set of virtually unknown manuscripts and fragments that are conserved in the Archivio Storico Diocesano in Cortona (I-CTd). The genesis of my discovery is the following. In the midst of doing research for the facsimile edition of the famous Laudario of Cortona, I was studying the handwriting of the scribe and searching for other instances of the same handwriting in liturgical books, when it occurred to me that it might be helpful to start by comparing the paleography of notated liturgical manuscripts from the same epoch. In the course of this investigation, I came across a great number of very interesting liturgical books.

From the overview of liturgical collections conducted by Giacomo Baroffio and published in his Iter liturgicum Italicum (2011), I extrapolated a preliminary list, comprising many manuscripts from the late thirteenth century in the areas of Tuscany and Umbria. The number of sources is vast: over six hundred in just a short time span and in a restricted area. Limiting the investigation to important churches within a radius of eighty kilometers from Cortona known to have renewed or copied ex novo the series of liturgical books results in the following list of twenty-one churches:

Churches and cathedrals Arezzo, Pieve of Santa Maria Grosseto, cathedral Gubbio, San Pietro Impruneta, Santa Maria Massa Marittima, cathedral Prato, Santa Maria dell’Umiltà (Chiesanuova) Siena, cathedral

Dominicans Arezzo, San Domenico (two antiphonaries in Arezzo – I-ARd –, two in the Municipal Library of Castiglion Fiorentino) Città di Castello, San Domenico Firenze, Santa Maria Novella Orvieto, San Domenico Perugia, San Domenico Spoleto, San Salvatore (now in Perugia, Biblioteca Augusta, I-PEc mss. 2790-2798)

49 CANTUS PLANUS 2014 GOZZI — MANUSCRIPTS IN CORTONA: FRAGMENTS AND LITURGICAL BOOKS

Franciscans The cycle of five Antiphonaries formed part of a single copying job carried out in 1331. Carmignano (Prato), San Francesco (now in Pistoia, Archivio Diocesano, I-PSad) The documents were copied by a scribe for the church of the Dominican convent Cortona, San Francesco of San Domenico during the rule of Ranieri Casali, Lord of Cortona, and they were Firenze, Santa Croce commissioned by the Cortona nobleman Martino Mattei, who was once married to Todi, San Fortunato a Genoese noblewoman named Alinor Salvago. We learn all of this from two large rubrics, one placed in front of fol. 1 of codex B and the other at the end of codex G, fol. Augustinians 193v, which both read as follows: Massa Marittima, Sant’Agostino Firenze, Santo Stefano al Ponte In Christi Dei nomine Ihesu, anno domini Millesimo CCC° XXX° primo, indictione XIIIIa, tempore domini Iohannis pape xxii. Hoc opus fieri fecit vir nobilis, sapiens et discretus Martinus magistri Mathei de Civitate Cortone, ad honorem Dei et gloriose Virginis Marie matris eius et Servites (Servants of Mary) beati Dominici patris et fundatoris ordinis fratrum Predicatorum, pro salute anime sue et patris et Firenze, Santissima Annunziata matris et domine Alinore de Salvaticis de Ianuis, uxoris olim dicti Martini, tempore honorabilis et Siena, Santa Maria nobilis militis domini Raynerii, generalis populi et Civitatis Cortone domini.

In the name of Jesus Christ, God, in the year of the Lord 1331, fourteenth indiction, in the time of Pope John xxii. The noble, wise and discreet Martino of master Matteo of the town of Cortona, had Seven important churches or cathedrals around Cortona were in possession of this work done, in honor of God, and of the glorious Virgin Mary, his mother, and of the Blessed liturgical books, as were six churches of Dominican convents (namely the Order of Dominicus, father and founder of the Preacher Fathers, for the salvation of his soul, his father’s Preachers), four Franciscan churches, two Augustinian churches, and two belonging and mother’s, and of Mrs. Alinor Salvago of Genoa, formerly wife of the said Martin, at the time of to the Servants of Mary. the honorable and noble captain Ranieri, the general of the people and lord of the city of Cortona.

The second half of the thirteen century was a period of great fervor in the foundation Frescucci provides an inaccurate translation of the text. For instance, he interprets «de of new convents of religious orders, especially among the Dominicans and the Ianuis» as a patronymic, something like ‘[daughter] of Janus’ (of John). Nevertheless, Franciscans. During the same years and further north, San Nicola in Pisa (Augustinians), his article contains an interesting photograph on page 23 showing that manuscript A San Francesco in Pisa (Franciscans), San Romano in Lucca (Dominicans), and San displayed exactly the same rubric, in the verse of the first folio. That rubric is now missing. Francesco al Prato in Pistoia (Franciscans) were also provided with choral books. The following is a list of the contents of the antiphonaries: This is an impressive list, yet it is clearly incomplete. Many manuscripts have been lost, and others are not listed in Baroffio’s catalogue. As a matter of fact, Baroffio I-CTd ms. A From Dominica i Adventus to Octava Epiphaniae mentions only one of the manuscripts I am going to present to you, and he lists it Sanctorale from Saint Andreas to Saints Innocentes under the heading ‘Biblioteca del Seminario’ (Baroffio 2011, p. 105), although it is I-CTd ms. G From Dominica i in Quadragesima to Sabbatum Sanctum no longer held there, and is instead kept with the other liturgical manuscripts in the Historical Diocesan Archive of Cortona (I-CTd). In addition to fourteen fragments, I-CTd ms. B From Sabbatum Sanctum to Festa Sanctissimae Trinitatis et Corporis Domini Commune sanctorum they amount to seven major manuscripts in all, including five antiphonaries and two graduals. These seven choral books are briefly described in I manoscritti medievali I-CTd ms. C From Dominica i post Trinitatem to Dominica i mensis septembris della Provincia di Arezzo. Cortona (Caldelli, et al. 2011, pp. 10-12, 99-103), whose From Saint Iohannes Baptista to Saint Catharina only bibliographic reference is to a 1976 article by Bruno Frescucci. Frescucci’s I-CTd ms. F From Dominica xiv post Trinitatem to Dominica xxv post Trinitatem article, however, is riddled with errors and inaccuracies; after all, it was just an amateurish description written by a priest for an antiques showroom catalog. Today these manuscripts are designated by alphabetical lettering from A to G, with the The content of the antiphonaries is complete, but the books are not currently arranged letters written on sheets of white paper placed at the beginning of each volume, as is in the form in which they were originally copied. In the second half of the seventeenth common practice in identifying manuscripts. However, these handwritten marks are century, a number of fascicles were moved, and the manuscripts were reassembled. certainly not the originals.

50 51 CANTUS PLANUS 2014 GOZZI — MANUSCRIPTS IN CORTONA: FRAGMENTS AND LITURGICAL BOOKS

For example, we can see that the volume containing the proper of the saints was split Cortona, Archivio Storico Diocesano, ms. D (I-CTd D) into two parts, now found separately in the current manuscripts A and C. fols. 1-170 Proprium de Tempore (from Dominica Resurrectionis to Dominica xxiii post Pentecosten) The cycle was intended to be richly illustrated. The Cortona, Archivio Storico Diocesano, fols. 170v-174 Dedicatio ecclesiae ms. A (I-CTd A), fol. 3v contains a blank space that was reserved for a large illuminated capital A for Aspiciens a longe, the first responsorium for the first Sunday of Advent. fols. 174-181 Missae votivae However, this ornamentation was never realized, and just a few initials were added by fols. 181v-183 Antiphonae ad aspersionem aquae (Vidi aquam, Asperges me) inexperienced hands, probably much later. Many initials were added a century after fols. 184-201 Kyriale the completion of the five large volumes. Most likely, the missing illumination reflects a lack of funds for completing the manuscripts. fols. 202-221 Sequentiae fols. 222-229 Missa Corporis Domini

Instead, the double set of twenty-one choral books now preserved in Perugia is richly fols. 230-234 Credo Regis, Sanctus with trope Divinum misterium decorated (Parmeggiani 2006). The oldest series, consisting of three graduals and five antiphonaries, is dated to the late thirteenth century. The second set consists of four graduals, eight antiphonaries and a proser, dating from the beginning of the fourteenth It is clear that at least one volume (containing the proprium and the commune of the century. The first group may come from the Dominican convent of Spoleto and the saints) is missing, and also that both volumes lack several sequences of the Dominican second from San Domenico in Perugia. These are now kept in the Biblioteca Comunale series. Augusta of Perugia (I-PEc: the city library) and are fully accessible for viewing online at http://augusta.alchimedia.com/ The fact that the mass of Corpus Domini is present but as a later addition proves that the manuscripts were written prior to 1317, when Pope John xxii extended the The cycle of Cortona antiphonaries has a clear Dominican structure. As we know, each feast of Corpus Domini to the entire Catholic Church (cf. Oppenheim and Toschi 1950 Dominican manuscript had to conform to the normative exemplar, a model established for a different date: 1314). The Solemnity of Corpus Christi was instituted in 1264 by in 1256 by Umberto De Romans (Humbert of Romans), the fifth master general of the Pope Urban iv as a feast on the Thursday after Pentecost for the entire Latin rite. The Order of Preachers (Boyle, Gy and Krupa 2004). traditional narrative has it that this act was inspired by a procession to Orvieto in 1263 after a village priest in Bolsena witnessed a Eucharistic miracle: while celebrating The series of antiphonaries was not executed by a single scribe. Codex A, for instance, mass, he broke the consecrated host and saw it bleed. Although this was the first shows different hands: from fol. 135 on, where the santorale starts, the copyist papally imposed universal feast for the Latin rite, it was not in fact widely celebrated changed. The same hand that copied the first few pages of ms. A can also be seen in the for half a century. It became a truly universal feast after the bull issued by Urban Commune sanctorum and other sections of ms. F (from fol. 69). The team of copyists, iv, but was promulgated only by Pope John xxii in 1317 (Rubin 1991, pp. 181-182; the scriptorium, seems to be different from that of the Perugian manuscripts, as there Walters 2006). The dating of the gradual is therefore a little earlier than the date of are no striking similarities in the handwriting. the antiphonaries, and can be established as the early years of the fourteenth century.

As for mss. D and E, they contain two graduals by the same copyist. Their content is I-CTd D contains the rubrics of the Sundays after Pentecost, and follows the Roman as follows: rite of counting Sundays starting from Pentecost as opposed to the Dominican rubric, which denominates Sundays as post Trinitatem Sundays. Nevertheless, the content Cortona, Archivio Storico Diocesano, ms. E (I-CTd E) was Dominican and followed the exemplar provided by Humbert of Romans. fols. 1-231 Proprium de Tempore (from Dominica i Adventus to Sabbatum Sanctum) Both volumes contain an addition from the fourteenth century (probably not much later fols. 232-234 Sequence Letabundus exultet fidelis chorus than their completion), the so-called Credo regis or Regis Siciliae, composed by Robert fols. 234-247 Kyriale of Anjou, King of Sicily. This Credo, together with the Credo Cardinalis, is the prototype of cantus fractus, or rhythmicized plainchant, a technique that became increasingly fols. 248-252 two Credo [Regis and Dominicalis] common from the fourteenth century forward (Gozzi 2012; Gozzi and Manganelli 2014).

52 53 CANTUS PLANUS 2014 GOZZI — MANUSCRIPTS IN CORTONA: FRAGMENTS AND LITURGICAL BOOKS

A complete set of sequences following the Dominican exemplar would come to twenty- In secundis autem vesperis antifona super psalmos Pie pater Dominice, questa antifona non ci sta, seven pieces. Of these, the codex Cortona, Diocesan Historical Archive, ms. D (I-CTd D) potrai perciò pigliar Pie pater della Messa, che sta nel libro F carte LXXI. contains the following seven: In secundis autem vesperis antifona super psalmos Pie pater Dominice, this antiphon is not available, so you can use Pie pater of the mass, which is in book F, folio LXXI. Victime paschali laudes (In resurrectione Domini), fol. 201 Omnes gentes plaudite (In ascensione Domini), fol. 203 Al Magnificat O lumen in libro A carte cento dieci a tergo. Sancti Spiritus adsit (In die Pentecosten), fol. 207 The Magnificat antiphon O lumen in book A, folio one hundred and ten on the back. Veni sancte Spiritus (Alia sequentia in Pentecosten), fol. 211 I-CTd ms. B, fol. 126v Profitentes unitatem (De Trinitate), fol. 213 Nota che l’Antifona Memoria la troverete nel libro V 5 e la memoria della domenica infra deto la Rex Salomon fecit templum (In dedicatione ecclesie), fol. 217 troverete nel libro V 74. Lauda Syon (interpolation) [Corpus Christi], fol. 224 Note that you will find the antiphon Memoria in book V, folio 5, and you will find the memory of Sunday infra the day aforementioned in book V, folio 74. The first sequence of the Dominican series, Letabundus, can be found on its own in codex E, in fol. 234, before the Kyrie and after the proprium de tempore that ends with Holy Saturday. The other, missing sequences must have been placed in a different The choir books, and in particular the antiphonaries, were reassembled in the second book, which is now lost. half of the seventeenth century. There is no known documentation for the reasons behind restructuring the Dominican books, but it was probably just to create a more The Dominicans arrived in Cortona in around 1240 and settled outside Porta San functional arrangement of the songs, for example, by putting the chants for the feasts Domenico (formerly Peccioverardi), but construction of the monastery was halted for of the saints together with the single volumes containing the proprium de tempore. lack of funds. The church was completed in 1314 and the convent in 1320. The original church building was demolished due to structural problems and likely to build the Since similar records as well as eye-catching ink foliation can also be found in the new church, now called San Domenico Vecchio, or old Saint Dominic. Built in the early series of manuscripts now preserved in Perugia, it is reasonable to assume that at fifteenth century, this church still stands. The books of Cortona probably remained in around this time (the late seventeenth century) the Dominican choral books in Perugia use until the eighteenth century and in fact show signs of heavy usage, despite their and Cortona were updated and rearranged by a single team of friars, who intervened lack of decoration and missing illuminated initials in many sections. In 1681, all of with rearrangements, corrections, and suggestions for using the various books. These the manuscripts were rebound and re-assembled. This is revealed by the numerous instructions may have been intended for a scribe who would then be responsible for marginal notes in Italian (some of them bearing a date) scattered throughout the executing a new manuscript, since in the seventeenth century handwriting was still in manuscripts. We find them mainly in the antiphonaries, such as: use for books of such dimensions, a feat that printing was not yet able to accomplish.

I-CTd ms. C, fol. 54v The same is likely to have occurred with all of the liturgical books with notation in Nota che il resto di questo responsorio ‘O crux gloriosa’ con tutte le altre antifone stanno nel libro the areas of Umbria and Tuscany, and perhaps in all of the other convents of the B a carte 55. Cerca come ti dico et non ti perdere d’animo et ivi troverai ancora in festo Michaeli et cetera. Dominican Order. Note that the rest of this responsory ‘O crux gloriosa’, along with all the other antiphons, are in book B, folio 55. Seek as I tell you and do not lose heart, and there you will still find In festo Michaeli et Figures 1a and 1b (pp. 56-57) show two different manuscripts from the series of Cortona cetera. and Perugia. Comparing their foliation, it is difficult to argue that number 178 was not written by the same hand, even though the two series have very little else in common. I-CTd ms. B, fol. 178 Nota in festo Translationis Patris Dominici cerca nel libro A carte 96. The first thing one notices is that the page numbering, a later addition in both the Note in festo translationis Patris Dominici look in book A, folio 96. Perugia and the Cortona books, were certainly written by the same hands. It seems Antifona super psalmos in primis vesperis Gaude felix cerca nel libro A carte 96 a tergo. Al that at least two hands were at work, and maybe more, unless the two annotators Magnificat antifona si dice Magne pater in libro A carte 101. changed pen or completed their task over a number of different sessions. The notes in Antiphon super psalmos in primis vesperis Gaude felix, look in book A, folio 96 on the back. For the the margins are also a shared feature in the two sets of manuscripts. In Perugia, one of Magnificatantiphon one sings Magne pater in book A, folio 101. them bears the date 1716 (Parmeggiani 2006, p. 143).

54 55 CANTUS PLANUS 2014 GOZZI — MANUSCRIPTS IN CORTONA: FRAGMENTS AND LITURGICAL BOOKS

Figure 1a. Cortona, Archivio storico diocesano, ms. B (I-CTd ms. B), fol. 178 Figure 1b. Perugia, Biblioteca comunale augusta, I-PEc ms. 2793, fol. 178

56 57 CANTUS PLANUS 2014 GOZZI — MANUSCRIPTS IN CORTONA: FRAGMENTS AND LITURGICAL BOOKS

The fragments The tenth fragment is probably a notated breviary (fourteenth-century) containing the Finally, I would like to briefly describe some of the still unpublished fragments in the feast of the Visitation of Mary (the fragments of text that can be read in it, however, Archivio Storico Diocesano di Cortona (I-CTd). I want to thank the archivist, Barbara lack correspondences in the repertoires). Due to its apparent originality, it might be Giappichelli, for supplying photos. interesting to examine this fragment, but as it is in very poor condition, it would require sophisticated equipment to decipher it. The first fragment belongs to a noted fifteenth-century breviary. The folio contains the first two responsories for the Feast of Corpus Christi: The eleventh fragment is from an antiphonary (fifteenth-century) with the Commune Bvm: responsory Felix namque es sacra virgo Maria et omni laude dignissima quia ex Responsorium i. [Immolabit haedum multitudo fili]orum Israel ad vesperam Paschae. te ortus est sol iustitiae. Responsory Virgo dei genitrix quem totus non capit orbis in tua Et edent carnes et azimos panes. V. Pascha nostrum immolatus est Christus itaque se clausit viscera factus homo (Cao 6333). epulemur in azymis sinceritatis et veritatis. Responsorium ii. Comedetis carnes et saturabimini panibus iste est panis quem dedit The twelfth fragment contains the tractus for Holy Saturday: Cantemus Domino vobis dominus ad ve[scendum]. gloriose, followed by the oratio and the prophetia v (Haec est hereditas) derived from a plenary missal from the fourteenth century. The second fragment belongs to a fifteenth-century gradual and contains the Missa votiva de Sancta Maria in Sabbato (graduale Benedicta et venerabilis, alleluia Post The thirteenth fragment contains two reponsories for the feast of Corpus Christi and partum, offertorium Ave Maria) and the Missa pro peccatis (introitus Misereris omnium, belongs to an antiphonary from the fourteenth century. alleluia De profundis). The fourteenth fragment contains the songs of the proprium missae for the second and The third fragment is from a fifteenth-century missal and contains some prefaces to third Sundays of Advent from a fourteenth-century gradual. the Sanctus. On the basis of the foregoing observations, it is my view that the Cortona manuscripts The fourth fragment belongs to the same gradual as fragment 2 (Missae votivae) and presented herein deserve to be rescued from their relative obscurity and taken duly served as the cover for a Register of Marriages 1540-1757. into consideration and researched. I-CTd mss. D and E consist of graduals, and were written by a single scribe before 1317, probably in the first years of the fourteenth The fifth fragment belongs to an antiphonary (possibly fifteenth-century) containing century. They present a Dominican format but occasionally employ Roman rubrics, the common of saints (we notice the unusual presence of a text belonging to the Secreta which is unusual for the liturgical sequence of this order. and Postcommunio in the verso). As we have seen, the homogeneous antiphonaries (I-CTd mss. A, B, C, F and G) The sixth fragment is an antiphonary (possibly fifteenth-century) containing a folio present a documented date (1331) and were written by several copyists following from Officium defunctorum. the Dominican exemplar. Although their format proves that they were intended to be illuminated, they were decorated with just a few bare capital letters and occasional The seventh fragment is an antiphonary (thirteenth/fourteenth-century) containing a ornamentation. The scribes of the antiphonaries are different from those of the Perugia section of the Officium of Saint Michael. manuscripts, which are also older and come from a much more organized and well- funded scriptorium. But comparison of the manuscripts with the well-known Perugia The eighth is the fragment of a gradual (fourteenth-century) containing the end of the corpus could lead to interesting discoveries and developments in the knowledge of Epiphany mass (offertorium Reges Tharsis and communio Vidimus stellam) and the this fascinating topic of Dominican liturgy.The order of both sets of manuscripts was Dominica infra Octavam Epiphaniae. rearranged during the second half of the seventeenth century and the beginning of the eighteenth century, and marginal notes were added. It would be valuable to investigate The ninth fragment, the oldest, and therefore rather rare, is a twelfth century these glossae, taking into consideration the similarity in content of both the Perugia antiphonary containing the Epiphany feast. and Cortona manuscripts.

58 59 CANTUS PLANUS 2014

Investigation of other manuscripts from this area between Tuscany and Umbria, as Karin Strinnholm Lagergren well as a broader foray into the Dominican manuscripts, may provide valuable insight into this phenomenon of rearrangement, and could also lead to other unforeseeable The Birgittine Abbey of Maria Refugie discoveries in a field which is considered so consistent and uniform as not to deserve Five hundred years of manuscript production much further attention. In spite of the fixed norm that has inspired the Dominican liturgical tradition in an unaltered form since 1256, we cannot help noticing that those documents are not so similar to each other after all, and further study of the single documents could present exciting and stimulating surprises for the scholar. The liturgical manuscripts from the Birgittine Abbey Maria Refugie in Uden, the Netherlands, are a demonstration of the fact that handwritten manuscripts continued to be produced even after the introduction of the printing press in Europe. Undertaking such an expensive and laborious process as book printing was probably out of the question for small liturgical communities, especially when they could continue producing books using the same method they had relied on for centuries. Here we discuss one such handwritten text preserved in the abbey library at Maria Refugie, where the manuscript tradition can be traced from ca. 1500 up to the nineteenth century. The abbey library at Maria Refugie belongs to the Order of Saint Birgitta of Sweden (ca. 1303-1373), and the abbey will be referred to hereinafter by Bibliography two different names: Mariënwater and Maria Refugie. The latter name was adopted Baroffio Giacomo (2011), Iter Liturgicum Italicum. Editio Maior, Stroncone: Associazione San Michele when the community moved to the city of Uden in 1713, as further described below. Arcangelo. The abbey and the sources used and produced there, including manuscripts held in Boyle Leonard E. - Gy Pierre-Marie - Krupa Pawels eds. (2004), Aux origines de la liturgie dominicaine: le other collections today, will be referred to as Uden and the Uden sources. manuscrit Santa Sabina XIV L 1, Roma: École Française de Rome.

Caldelli Elisabetta - Gallori Francesca - Pantarotto Martina - Parigi Maria Cristina - Pomaro Gabriella - The very first Birgittine abbey was founded at Vadstena (Sweden) in 1384. The Stoppacci Patrizia (2011), I manoscritti medievali della Provincia di Arezzo. Cortona, Firenze: Sismel- abbey, which became wealthy through donations, testaments and dowries, was Edizioni del Galluzzo (Biblioteche e Archivi 25). Sweden’s major land owner in the late Middle Ages, until it closed in 1595 due to the Frescucci Bruno (1976), Miniatura a Cortona in un’opera del 1331 in Città di Cortona. Palazzo Vagnotti, Reformation. Outside Sweden, the foundress, Saint Birgitta, is mainly known as a XIV mostra-mercato nazionale del mobile antico (29 agosto - 26 settembre 1976), Firenze: Alinari mystic. The fact that she herself took an active part in the creation of her Order and Baglioni, pp. 21-50. its liturgy is perhaps less well known. An ideal Birgittine abbey was designed as a Gozzi Marco (2012), I prototipi del canto fratto: Credo regis e Credo cardinalis in Marco Gozzi ed., Cantus double abbey for a maximum of sixty nuns and thirteen friars. The friars’ primary fractus italiano: un’antologia, Hildesheim-Zürich-New York: Olms (Musica mensurabilis 4), pp. 137-154. task was to assist the sisters with the celebrations of mass and to hear confessions. Gozzi Marco - Manganelli Michele (2014), Un nuovo frammento italiano del Trecento: il Manoscritto M The Order was particularly devoted to the Virgin Mary in her role as Mother of the 50 della Biblioteca Michele Manganelli in Marco Gozzi - Agostino Ziino - Francesco Zimei eds., L’ars Redeemer. nova italiana del Trecento VIII: Beyond 50 Years of Ars Nova Studies at Certaldo 1959-2009, Lucca: Lim, pp. 183-216. The Birgittine Order spread to the continent in the fifteenth century, and became Oppenheim Filippo - Toschi Paolo (1950), Corpus Domini, festa in Enciclopedia Cattolica, 12 vols., Città del Vaticano: Ente per l’Enciclopedia Cattolica e per il Libro Cattolico, vol. 4, pp. 611-614. particularly popular in the Low Countries and Lower Rhine area. Mariënwater, founded in ca. 1434, was the first Birgittine abbey in this area. During the period Parmeggiani Claudia ed. (2006), Canto e colore: i corali di San Domenico di Perugia nella Biblioteca comunale 1437-1482, other double abbeys were founded there as well, with Mariënwater as Augusta, 13.-14. sec.: 11 marzo-17 aprile 2006, Perugia, Sala Lippi, Unicredit banca, Perugia: Volumnia. their mother abbey. Rubin Miri (1991), Corpus Christi: The Eucharist in Late Medieval Culture, Cambridge: Cambridge University Press, pp. 181-182.

Walters Barbara R. (2006), The Feast of Corpus Christi, University Park, PA: Pennsylvania State University The paper is a report from the research project The transmission of the office Cantus Sororum in the Low Countries carried Press. out at KU Leuven, Belgium, during 2013-2015.

60 61 CANTUS PLANUS 2014 STRINNHOLM LAGERGREN — THE BIRGITTINE ABBEY OF MARIA REFUGIE

The Birgittine liturgy was a mix of well-known elements from the Roman Catholic Table 1. Graduals and antiphonals produced or used at Mariënwater and Maria Refugie (Manuscript names and numbers as in Sander Olsen 2002) common stock, adaptations of text and music, and unique contributions. Mass was celebrated according to the customs of the local diocese, meaning that the liturgy changed depending on the diocese in which the abbey was situated. The daily mass Type Title Shelf number Dating for the nuns in all the abbeys was the mass of Our Lady, the Salve sancta parens, Period 1, ca. 1480-ca. 1510 while a number of other mass formulas were used for Marian feasts. So far, this is commonplace Roman Catholic practice. The liturgy of the nuns’ office, which was Antiphoner- Officia et Missae de Bmv NL-DHk KB 71 A 21 ca. 1500 gradual (title after Boeren et al. 1988) based on the popular Officium parvum beate Marie Virginis and adopted in all of the Antiphoner- Birgittine abbeys, took an unusual departure. In the hands of Saint Birgitta and her Birgittanum Uden HS K:An 1 gradual collaborator and confessor, Petrus Olavi of Skänninge (who died in 1378), this office Antiphoner Antiphonale Birgittanum Uden HS K:An 2 ca. 1500 was extended into an officium hebdomadarium, with a specific theme for each day of Antiphoner Antiphonale Birgittanum Uden HS K:An 3 ca. 1500 the week, all celebrating one particular aspect of the Virgin Mary. The texts and music were to a large extent borrowed from other sources, but the office also contained Antiphoner Antiphonale Birgittanum Uden HS K:An 4 ca. 1500 newly composed melodies and texts. These unique contributions were the result of Antiphoner Antiphonale Birgittanum Uden MRK 072 ca. 1500 collaboration between Saint Birgitta (who was particularly responsible for the texts Antiphoner Antiphonale Birgittanum Uden HS K:An 4 16th century for the nocturne readings, the so-called Sermo Angelicus) and Petrus Olavi. Together, Period 2, ca. 1639-ca. 1660 the liturgies for the friars and the nuns made up the so-called greater liturgy, a term coined by the Swedish musicologist Ingmar Milveden. Gradual Graduale Birgittanum Uden HS K:Gr 1 1640 Antiphoner Antiphonale Birgittaum Uden HS K:An 7 1640 The manuscripts Gradual Graduale Birgittanum Uden HS K:Gr 2 1641 The Uden manuscripts only consist of books used by nuns. Manuscripts employed Gradual Graduale Birgittanum Uden HS K:Gr 3 1644 by the friars are unknown, most probably because the Brethren took all their books Gradual Graduale Birgittanum Uden HS K:Gr 4 1644 with them at the dissolution of the double abbey in 1652. To this day, they have not Gradual Graduale Birgittanum Uden HS K:Gr 6 ca. 1645 been found. The conclusions presented here are based on the fifty-seven preserved Antiphoner Antiphonale Birgittanum Uden HS K:An 9 1646 antiphoners and graduals presented in table 1. The choice of antiphoners and graduals Antiphoner Antiphonale Birgittanum Uden HS K:An 10 1646 is motivated by the fact that these are the only kind of notated musical manuscripts in which we can follow the liturgical and musical tradition from ca. 1500 up to the Antiphoner Antiphonale Birgittanum Uden HS K:An 14 ca. 1647 nineteenth century. Hymnals and processionals are thus not considered. Gradual Graduale Birgittanum Uden HS K:Gr 5 1648 Antiphoner Antiphonale Birgittanum Uden HS K:An 12 1648 Manuscript production of the Uden sources falls into four periods: Antiphoner Antiphonale Birgittanum Uden HS K:An 13 ca. 1648 Antiphoner Antiphonale Birgittanum Uden HS K:An 11 1649 Period 1: ca. 1480 - ca. 1510 Gradual Graduale Birgittanum Uden HS K:Gr 7 ca. 1650 Period 2: ca. 1639 - ca. 1660 Antiphoner Antiphonale Birgittanum Uden HS K:An 6 ca. 1650 Period 3: ca. 1728 - ca. 1760 Gradual Graduale Birgittanum Uden HS K:Gr 8 1650 Period 4: ca. 1843 - after 1883 (?) Gradual Gradual Graduale Birgittanum Uden HS K:Gr 12 ca. 1650 Antiphoner Antiphonarium Birgittanum Weert HS 2 Mariënwater ca. 1650 Table 1 presents the manuscripts chronologically by year or approximate dating, but Gradual Antiphonarium Birgittanum Weert HS 3 Mariënwater ca. 1650 no attempt has been made to date the manuscripts written in the same year in relation to each other. The establishment of the periods is based on the first and last dates Gradual Graduale Birgittanum Uden HS K:Gr 14 ca. 1650 found in the manuscripts, wherever a date was possible to establish. Otherwise, date Gradual Graduale Birgittanum Uden HS K:Gr 10 1651 assignments are based on the style of script, decoration and initials. Gradual Graduale Birgittanum Uden HS K:Gr 11 1655 Gradual Graduale Birgittanum Uden HS K:Gr 9 17th century

62 63 CANTUS PLANUS 2014 STRINNHOLM LAGERGREN — THE BIRGITTINE ABBEY OF MARIA REFUGIE

Period 3, ca. 1720-ca. 1760 Period 1, ca. 1480-ca. 1510 No notated liturgical manuscripts are preserved from the first decades of the abbey, but Gradual Graduale Marianum Birgittanum Uden HS K:Gr 16 1728 a number of both liturgical (i.e., breviaries) and non-liturgical (i.e., the revelations of Gradual Graduale Marianum Birgittanum Uden HS K:Gr 17 1728 Saint Birgitta) books have been preserved. The notated liturgical manuscripts date from Gradual Graduale Marianum Birgittanum Weert HS 7 1729 ca. 1500. The abbey had a scriptorium of high quality at this time under the Abbess Gradual Graduale Marianum Birgittanum Weert HS 9 1730 Heylwich Oeyuers (1460-1503, died ca. 1520), reaching its heyday ca. 1480-1510. Antiphoner Antiphonarium Birgittanum Weert HS 6 Mariënwater 1732 The Antiphoner-Gradual Officia et missae de Bmv, NL-DHk KB 71 A 21 Antiphoner- Antiphonale and Graduale Uden HS K:An 16 1735 gradual Birgittanum One of the books written during this heyday is the antiphoner-gradual NL-DHk KB Graduale Antiphonale Mariano Birgittanum Uden HS K:An 17 1735 71 A 21, kept today in the Royal Library in The Hague. This manuscript is one of the three extant antiphoner-graduals from Uden (see table 1). The fact that there are Antiphoner Antiphonale Mariano Birgittanum Uden HS K:An 18 1736 so few of these combined liturgical volumes in comparison to the other manuscripts Antiphoner Antiphonale Mariano Birgittanum Uden HS K:An 19 ca. 1740 makes it likely that the antiphoner-graduals held a special place in the liturgy. I argue Antiphoner Antiphonale Mariano Birgittanum Uden HS K:An 20 ca. 1740 that they belonged to the chorista (the sister in charge of the singing, as specified in Antiphoner Antiphonale Mariano Birgittanum Uden HS K:An 21 1743 the Birgittine Regula), although no inscriptions can confirm this. Like Uden HS K:An Antiphoner Antiphonale Mariano Birgittanum Uden HS K:An 22 ca. 1750 16, the antiphoner-gradual 71 A 21 measures 30×40 cm, which is unusually large in comparison to the other preserved manuscripts, but a good size to be placed on Graduale Graduale Birgittanum Uden HS K:Gr 20 ca. 1760 a music stand. The manuscripts are generally much smaller, more appropriate for Graduale Graduale Birgittanum Uden HS K:Gr 18 18th century holding in one’s hands. Of the Uden sources, 71 A 21 is also the only one that contains Graduale Graduale Birgittanum Uden HS K:Gr 19 18th century the complete readings for matins and, in many cases, the complete psalm texts as Antiphoner Antiphonale Birgittanum Uden HS K:An 23 18th century opposed to just the incipit. Additions in the margins of new Gloria Patri formulas in Graduale Graduale Birgittanum Uden HS K:Gr 21 18th century the great responsories point to use of the 71 A 21 into the seventeenth century. Also, during the seventeenth century, several melodies were reworked, particularly the Graduale Graduale Birgittanum Uden HS K:Gr 22 18th century clivis and podatus, which were reduced to a single note. It is worth pointing out that 71 Graduale Graduale Birgittanum Uden HS K:Gr 23 18th century a 21 is the only extant manuscript among the Uden sources that is entirely written on Graduale Graduale Birgittanum Uden HS K:Gr 24 18th century parchment. It could be that this copy was produced as an archetype for the Birgittine Graduale Graduale Birgittanum Uden HS K:Gr 25 18th century sisters’ liturgy, but it has not yet been possible to establish the relation between this Graduale [Graduale Mariano Birgittanum] Weert HS 5 18th century and the other manuscripts.

Period 4, ca. 1843-after 1883? After the first scribal period, conditions gradually became more difficult for the abbey, mainly because of the religious wars. After the iconoclastic riots of 1566, the nuns and Gradual Graduale Mariano Birgittanum Uden HS K:Gr 26 1843 the friars moved out of the abbey buildings and went to live with families in the nearby Antiphoner Antiphonale Mariano Birgittanum Uden HS K:An 24 1846 city of ’s-Hertogenbosch for several decades. We do not know exactly when, but by Antiphoner Antiphonale Birgittanum Uden HS K:An 25 19th century 1608 at the latest, the community had returned to Mariënwater, and a new era began. Antiphoner Antiphonale Birgittanum Uden HS K:An 26 19th century Period 2, ca. 1639 - ca. 1660 After about twenty years of re-consolidation, a period of intense scribal activity Very few manuscripts were produced in the gaps between these four periods, and commenced. Since many manuscripts date from this time, we have a good idea of only one of them is a notated liturgical book (Uden HS K:An 5). As will become clear what was being produced. from the following text, the reason for this is most likely the difficult and even chaotic times experienced during those lapses in time. A short survey of the periods of scribal During this period, the community received many vocations, thus creating a demand production will now follow. for new liturgical manuscripts. In 1639, the abbey also received fourteen nuns from

64 65 CANTUS PLANUS 2014 STRINNHOLM LAGERGREN — THE BIRGITTINE ABBEY OF MARIA REFUGIE their daughter foundation, Mariëntroon at Dendermonde. Due to a dispute with the The community at Mariënwater continued to function, but now left only to the nuns. bishop, Anthonius Triest of Ghent, the abbess and thirteen nuns fled to Mariënwater The situation grew increasingly intolerable for the Birgittine sisters and other religious and stayed there until they were summoned back to Mariëntroon in 1650. One of the communities in the area, eventually causing them to move from Mariënwater to the occupations that these sisters were engaged in while in Mariënwater was copying Catholic Duchy of Ravenstein, a Catholic enclave located between two Protestant books, as indicated by anonymous inscriptions, such as «written by a refugee sister regions in the Netherlands. Several abbeys, monasteries and convents relocated to this from Mariëntroon». area after the peace of Westphalia. In 1710, the Birgittine sisters purchased an old abbey building from the Crosiers (Kruisheren), and in 1713 they officially established It was not only the content that was faithfully copied from period 1, but also the their community in the renovated abbey buildings. The sisters called their new home historiated initials in late medieval style, which, however, were now produced in Maria Refugie (the Refuge of the Virgin Mary’s). As in the early seventeenth century, cruder fashion: the heyday of the scriptorium was definitely over. they had to rebuild the community once again, this time apart from the community of brothers, although they did have Birgittine brethren from their daughter foundations There are no antiphoner-graduals from period 2 preserved, and all of the manuscripts assist them at mass and confession. from period 2 are smaller, measuring on average about 17×10 cm. The liturgy is the same as in period 1 for the office and mass formulas, but there are signs that the melodic Period 3, ca. 1720-ca. 1760 repertoire is beginning to change. Some manuscripts have new melodies pasted in It was a much diminished community that moved to Maria Refugie: just four nuns, over the original ones, and in others the music has been erased and replaced with including the Abbess. To enlarge the community, they asked for help from their new melodies, which are reworkings of the original melodies. This will be discussed daughter foundation Marienblum in Kalkar, which sent a small number of nuns to further under period 3. join them. After a period of stabilization, production of new manuscripts was taken up again. Manuscripts from this period differ in two important ways from those produced Outside circumstances also affected the temporary additions of chants. In addition to in periods 1 and 2: the style of decoration changes – the historiated initials disappear armed conflicts, seventeenth-century Europe had to contend with repeat epidemics and are replaced by a more up-to-date design in late baroque or even rococo style. of contagious diseases, including the plague. The Low Countries and Mariënwater were some of the areas severely affected. One example, an epidemic in the 1630s that The graduals are now of three kinds. The first one (A) includes graduals for the same devastated the convent with casualties, is reflected in the liturgy: a Marian antiphon feasts as in periods 1 and 2: against the plague – Hec est preclarum vas – was sung after the nun’s daily mass of Our

Lady during the late seventeenth century and in the early eighteeenth century. This A. Graduals for the same feasts as in periods 1 and 2 antiphon was not unique to the Birgittines, and in fact was a popular antiphon against plague and other contagious diseases in the Low Countries from the late Middle Ages Daily Mass of Our Lady onwards. In the Uden graduals, it is only found as additions in manuscripts from Advent Period 2, but during the eighteenth century it found its way into the private prayer Christmastide books of the Uden Sisters. Christmas until PurificatioB mv From Period 2 to Period 3 Presentatio Bmv One event that would have a profound impact on the Birgittine community at Mariënwater was the Peace of Westphalia in 1648, which signalled the end of the Assumptio Bmv

Thirty Years’ War. The peace treaty fixed the borders between Protestant and Catholic Visitatio Bmv territories, and the result was that Mariënwater now belonged to the Protestant part. Nativitas Bmv The nuns were allowed to remain at their abbey, but they were not allowed to take novices, and all male clergy had to leave the territory. Moreover, by this time, the Conceptio Bmv Catholic Church had become suspicious of double abbeys, which eventually led to the Presentatio Bmv dissolution of all Birgittine double abbeys in the Low Countries during the seventeeenth century. In 1652 the friars at Mariënwater moved to Hoboken, near Antwerp. Purificatio Bmv

66 67 CANTUS PLANUS 2014 STRINNHOLM LAGERGREN — THE BIRGITTINE ABBEY OF MARIA REFUGIE

In addition to these, the large Birgittine repertoire of seventeen sequences continued office, seven of the antiphons, hymns, invitatory antiphons, short responsories and being used as they are found in the earliest Birgittine sources from the fifteenth century. great responsories were reworked. The melodies are clearly modelled on the originals In this respect, the Birgittines did not comply with the Council of Trent’s decree to found in period 1 and period 2, but the melismas are often shortened or moved to abolish the use of sequences from the mass. other nearby syllables, and the melodic contours changed. Importance is given to the correct accentuation of the words, and sometimes the mode was also changed. The next type of graduals (B) are those that only contain new feasts (compared to the It is difficult to say exactly what was achieved by these reworkings, but the aim may list above) and are of a general character. They lack the Marian characteristics found have been to enhance the mode in question, to emphasise the text, and to conform to in the graduals above. The new feasts are: changes in musical taste in accordance with the musical perception of the Baroque and the wishes of the Council of Trent. No documentation from the Birgittines at Uden B. New feasts has been found that can shed light on this procedure, but it can be concluded from Dominica quarta in Quadragesima the manuscripts that these reworkings took place over a long period, which probably started well before the beginning of period 3. This assumption is based on the fact Feria quinta in cena Domini that some of the reworked melodies were written by the first scribe, while others Dominica Resurrectionis are written on top of the erased first melody. Exactly which reworked melodies were Ascensio written into the manuscripts as original melodies and which were written on top of Dominica Pentecostes erased staves remains to be investigated. The tradition of the texts is, on the other hand, is entirely stable. In festo trinitatis et in professionibus1 Festum corporis Christi One possible reason behind these melody reworkings is that the new sisters who came Saint Birgitta from Marienblum in 1700 brought with them manuscripts that contained these altered Saint Catherine of Alexandria melodies, and that they were then copied into the newly produced manuscripts in Uden. Pro defunctis One of the Marienblum sisters, Theodora Alexia de Haen, was elected Mother Superior of the abbey in 1705 and it may be she who carried out this liturgical reform. Since no Saint Katarina of Vadstena manuscripts from Marienblum dating from before the eighteenth century are known, 1 this hypothesis cannot be confirmed.2 What we do know is that at the beginning of the The third type of graduals (C) combine types A and B, to form a larger corpus that is eighteenth century, the Birgittine sisters at Uden made a new start in many ways. The well adapted to the liturgical year, with a special emphasis on Marian feasts. community and the liturgy had both suffered losses after the departure of the brethren. The sisters managed to rebuild their community in a new location and under a new My conclusion about the reason for this development is that the sisters took over the name, with new members joining them from another Birgittine community. All of masses from the liturgy of the brothers’ following dissolution of the double abbey in these factors may have led to greater receptiveness to new liturgical usages. 1652. If this assumption is correct, it provides proof that the sisters’ liturgy could be adapted to the liturgical needs after the double abbey had split. Whether this actually Period 4, 1843-1883 did happen in the dissolved abbeys has been a matter of dispute; however, we can now 1784 is an emblematic year in the history of the Low Countries, marking the start of say that in Uden, at least, it really seems to have been the case, although we do not secularization with the suppression of monasteries and churches by Emperor Joseph know whether this was a general trend among the Birgittines. II. Contemplative orders, such as the Birgittines, were the prime target. The Birgittine brotherhood disappeared as did all of the other Birgittines houses – both sisters and As for the musical content, a great number of the office melodies are reworked in brethren – in the Low Countries and in the Lower Rhine. The one exception was Marie the period 3 manuscripts. In fact, a consistent reworking of a large part of the corpus Refugie, which was the only Birgittine abbey to survive the Napoleonic era in this took place in this period. For example, of the twenty-four melodies of the Tuesday region. However, Marie Refugie was permitted to exist, and, as during the intervals between periods 1 and 2, the sisters were not allowed to receive any novices, although

1. This feast probably refers to the monastic profession. See Regula salvatoris, capitulum XI, paragraph 117. Latin text accessible as pdf-files through the website of the National Archives in Stockholm: Corpus Reuelacionum Sancte Birgitte (CRB), http://riksarkivet.se/crb, accessed 13 January 2015. 2. Three hymnals are contained in the Uden collection, but the earliest dates from 1733.

68 69 CANTUS PLANUS 2014 STRINNHOLM LAGERGREN — THE BIRGITTINE ABBEY OF MARIA REFUGIE a number of women who were helping them at the abbey made vows in secret. When activity at the abbey can be divided into four periods, each preceded by a period of monastic activity was reinstated in 1840, the secret vows made by those women change/crisis in abbey life. In the seventeenth century additions and changes in the meant that Maria Refugie suddenly had fifteen new sisters ready to make their solemn manuscripts provide clues to changed musical tradition . In the manuscripts from profession in 1840, ushering in a new era with a larger community than there had period 3, many of the office melodies have been reworked, while the text tradition been for a long time. In fact, there were so many sisters that Maria Refugie was able to remains stable. The sisters’ mass liturgy was enlarged, as they probably took on the establish a new Abbey in 1843, the first daughter foundation since the fifteenth century. feasts originally celebrated by the friars, who had to leave the abbey in 1652. A number This was the Maria Hart in Weert, also still in existence, although it now belongs to the of new directions seems to have emerged between periods 2 and 3 and resulted in new Hesselbladh Branch, a reformed branch of the Birgittine order. styles of manuscripts and reworked musical repertoires for the music of the office. On a larger scale, the abbey library of the Birgittines at Uden provides a tool for Only a small number of new manuscripts were produced at this time, and the nineteenth investigating long term uses and developments in liturgical music, ranging from century marks the end of the era of handwriting and the transition to printed material. the Late Middle Ages onwards, including the by and large unwritten history of The sisters at Weert acquired bookmaking equipment, and their first print was made post-medieval chant. In this respect, it can be seen as a case study of one particular in 1856: a processional for the sisters at Uden. In 1881 the sisters at Weert printed environment in a longue durée perspective, which makes it possible to study how one an antiphoner without vespers, which was printed in 1883. Handwritten manuscripts particular context shaped a repertoire. for the mass most probably continued to be used alongside this printed antiphoner, since no graduals for the sisters ever were printed. It is also possible that handwritten manuscripts were written with the printed books as their model. The manuscripts Uden HS K:An 25 and 26 are not dated, but were possibly written after 1883, since they seem to imitate the prints from 1881 and 1883. If this can be established, the era of handwriting lasted until at least 1883.

For obvious reasons, I have only discussed the manuscripts that have survived to this day. There is reason to believe that books have disappeared over the centuries due to the many trials the sisters and the brethren underwent. Exactly what has been lost we do not know, since no catalogue of the library existed or has been preserved since before the 1970s, when Ulla Sander Olsen compiled an unpublished catalogue. This was later published in 2002 in a revised version.3 What we do know is that the sisters sold some manuscripts in the nineteenth century to raise money for renovation of their abbey. For example, 71 A 21, which has already been mentioned, was bought by the Royal library in The Hague in 1837 at an auction of the estate of doctor Wellens in Rotterdam, who had acquired the manuscript previously from the sisters. Books and Bibliography sculptures from Uden are known to have been sold at least in 1860 and in 1875. Some Corpus Reuelacionum Sancte Birgitte (crb), http://riksarkivet.se/crb of these manuscripts and sculptures have now been identified and found in other (accessed 13 January 2015). collections or brought back to Uden, where some are now housed at the Museum for Milveden Ingmar (1973), Sjungen ödmjukhet in Andreas Lindblom ed., Vadstena klosters öden. Till Religious Art located in a part of the abbey building. 600-årsminnet av Birgittas död, Finspång: Finspångs bokhandel, pp. 145-159.

Conclusion Nilsson Ann-Marie (1991), On liturgical hymn melodies in Sweden during the Middle Ages, Ph. D. dissertation, Gothenburg University. The abbey library at Uden represents a monastic musical tradition that covers 500 years. The core repertoire of the sisters’ office (the Cantus sororum) and the daily Sander-Olsen Ulla (2002), Bibliotheca Birgittina Birgittinessenabdij Mariënwater, Maria Refugie: Uden N.Br.: gesticht ca. 1437 - overgeplaatst naar Uden 1713, Brussel: Algemeen Rijksarchief. mass of Our Lady (the Salve sancta parens) remained stable over the centuries. Scribal Sander-Olsen Ulla - Nyberg Tore - Sloth Carlsen Per eds. (2013), Birgitta Atlas. Saint Birgitta’s Monasteries. Die Klöster der Heligen Birgitta. A Transeuropean Project – Ein transeuropäisches Projekt, 3. All Birgittine manuscripts and prints referred to are listed in this catalogue. Uden: Societas Birgitta Europa.

70 71 Santiago Ruiz Torres - Juan Pablo Rubio Sadia

Liturgical fragments of the diocese of Sigüenza (eleventh-sixteenth centuries)

This article outlines the main features of a corpus of liturgical fragments from the eleventh to the sixteenth centuries that were uncovered in various archives of the diocese of Sigüenza-Guadalajara.1 The definitive results of this analysis have been recently published in the journal Medievalia (Ruiz Torres and Rubio Sadia 2017). Here, after presenting an overview of the major historical milestones in this geographical area, we shall introduce this collection of fragmenta codicum and discuss their dating and typology. We shall then focus on the two most striking exemplars: an antiphonary from around 1100 and a polyphonic manuscript dating back to the early fourteenth century.

Sigüenza is a small town in the north-eastern portion of the province of Guadalajara (Castilla-La Mancha), about 125 km from Madrid. Still a bishopric, its origins date back to the Visigothic era (Mansilla 1994, 1, pp. 294-295). After the Arab invasion of 711, Sigüenza’s strategic position led to it becoming one of numerous fortified towns along the eastern boundary of the frontier territory of the Marca Media in Al-Andalus (Herrera 1985, pp. 11-14). The city later fell into gradual decline, dwindling to a tiny village under the control of Medinaceli (Blázquez 1988, pp. 42-49). Some historians argue that a small Mozarabic community may have survived until the Reconquest, citing the existence of two churches located in the upper part of Sigüenza as evidence (Rubio Sadia 2011, pp. 92-93). However, no trace of Hispanic liturgy has emerged from the research we carried out.

The reconquest and restoration of the bishopric of Sigüenza occurred between 1121 and 1124 (Reilly 1982, pp. 177-179). Bernardo de Sédirac, Cluniac archbishop in Toledo (1086-1125), consecrated Bernardo de Agen as bishop of Sigüenza in 1121. Since the town was located at the frontier with the kingdom of Aragon, Bernardo de Sédirac was interested in establishing a suffragan see there in order to thwart expansion of the ecclesiastical province of Tarragona (ibidem, pp. 246-247). The Medieval diocese bordered on the diocese of Burgo de Osma to the north, on the Albarracín and Cuenca dioceses to the south, on those of Tarazona and Zaragoza to the east, and on the Segovia and Toledo bishoprics to the west.

1. This publication is part of the activities carried out by the research group ‘El canto llano en la época de la polifonía’ (HAR2010-17398) at the Universidad Complutense de Madrid, directed by Carmen Julia Gutiérrez.

73 CANTUS PLANUS 2014 RUIZ TORRES - RUBIO SADIA — LITURGICAL FRAGMENTS OF THE DIOCESE OF SIGÜENZA

The French origin of the first bishops in Sigüenza throughout the twelfth century is (Minguella 1910, 1, pp. 72-75, 244 and 375, no. xxiii). Additional documents allow us a key factor for understanding the configuration of the local liturgy. The first three to identify other clergymen as French in origin; these include a scribe’s diploma from quarters of the century saw the succession of three prelates from Toledo: the above- 1135 (ibidem, p. 356, no. viii), indicating that the owner was from Auch. The canon mentioned Bernardo de Agen (1121-1151), Pedro de Leucate (1152-1156), and community celebrated the liturgy in rather humble surroundings in Santa María la Cerebruno de Poitiers (1156-1166). These three bishops were responsible for the early Vieja until construction of the proto-gothic cathedral, with its clear Languedocien organization of the diocese and for the supply of liturgical books of the Roman Rite. influence and role of temple-fortress, was concluded during the last thirty years of the Of the three, the most notable is undoubtedly Bernardo. Hailing from Agen (Gascony), twelfth century. The earliest chapels of the cathedral were dedicated to Saint John the he probably came to Spain aiming to strengthen the canonry of Toledo (Ximenez de Baptist, Saint Augustine, Saints Peter and Paul, and Saint Thomas from Canterbury; Rada 1987, p. 210). He turned out to be an influential figure both in the canonry and at the central altar was devoted to Saint Mary, to whom the cathedral was dedicated. the royal court. It is therefore not surprising that he was nominated in 1151 by Alfonso Later, other chapels (Muñoz 1987, pp. 185-274) were added and dedicated to Saint vii for the Compostelan see, where he remained for approximately one year (Fletcher Valerius (thirteenth century), Saint Catherine, and Saint Liberata (fourteenth century), 1978, pp. 57-58). Although it is widely held by historians that Bernardo was a Cluniac Saint Michael, Saint Antoninus, and Corpus Christi (fifteenth century). Moreover, we (Minguella 1910, 1, p. 62; Martínez 1985, p. 92), there is no documentary evidence that know that at the end of the fifteenth century the city was organized around the two this is the case. What we do know is that he served as cantor at the Church of Toledo, parishes (the aforementioned Santiago and San Vicente) that had been built inside the the same office Cluniac Geraldo de Moissac had held before moving to Braga. The walls (García and Pérez 2009, 2, pp. 776-789), as well as the one inside the cathedral, presence of such a strong of these Franks in several Spanish dioceses sheds light on devoted to Saint Peter (Martínez 1990, p. 174). Another important settlement in the the transmission of the melodic repertoire to the local clergy, which was one of the diocese was Atienza, located about thirty km northeast of Sigüenza. Reconquered by main components in the transition to the Roman rite. the Aragon king in 1112, it already had eight or ten churches in 1200, including Saint Mary, Saint Bartholomew, the Holy Trinity, and Saint Giles (Layna Serrano 1945, pp. Also worth mentioning is the relation between Bishop Bernardo and the introduction 437-445; García and Pérez 2009, 1, pp. 171-216). of the worship of two southern-French saints: Liberata, known as Librada in Spain, and Sacerdos. During the twelfth century, Liberata’s remains were enshrined in the Two important facts reveal the significant role played by the diocese of Sigüenza in cathedral, and she was eventually recognized as the patron saint of Sigüenza, its church the Iberian Peninsula in the two centuries that followed. The first is that four of its and the diocese. The transfer of her relics from Aquitaine, together with those of Saint fifteenth-century prelates, Pedro de Fonseca, Alfonso Carrillo de Albornoz, Juan de Sacerdos, bishop of Limoges, was likely promoted by Bernardo himself (García 2004, Mella, and Pedro González de Mendoza (Ubieto 1989, pp. 363-364; Olea Álvarez 2014, pp. 201-211; Olea Álvarez 2011, p. 32). 1, pp. 135-190 and 221-307) also served as cardinals. The second is the founding of the city’s university in 1489. After Bernardo was assigned to Compostela in 1151, Don Pedro took his place in Sigüenza (Minguella 1910, 1, pp. 95-106). Hailing from Leucate (in the province of Our search for liturgical fragments in the diocese of Sigüenza involved three main Septimania) and a cousin of Bernardo, Don Pedro was likely serving as a prior in the archives: the Archive of the Cathedral (including documents in Medinaceli, a canonry of Sigüenza in 1150. Despite his brief pontificate of just four years, it is to him municipality assigned to the diocese of Sigüenza until 1956), the Parish House of San that we owe the raising of the first funds for the construction of the Romanic cathedral. Juan Bautista in Atienza, and the Provincial Historical Archive of Guadalajara. In total, Succeeding Don Pedro as bishop was Don Cerebruno, possibly born in Poitiers, who we found forty-four liturgical fragments dating from the eleventh to the sixteenth served as archdeacon of Toledo between 1139 and 1155 (Hernández 1996, pp. 69, 178 century, all predating the adoption of the Tridentine liturgy. Thirty-four of these forty- and 180; Minguella 1910, 1, pp. 106-114). Documentary evidence places him as prelate four fragments belong to the Archive of the Cathedral, two of them to Medinaceli, one to of Sigüenza from November 1156 to 25 October 1166; later, he was assigned to Toledo, Atienza, and the remaining seven to the Provincial Historical Archive of Guadalajara. where he remained until 1180. Don Cerebruno promoted construction of the Santiago All of the manuscripts are of plainsong, with the exception of a fragmentary bifolio and San Vincente parishes in the upper part of the city, as well as fortification of upper containing polyphonic music from the early fourteenth century. Sigüenza and construction of a new cathedral (Martínez 1990, p. 154). Once the documents were categorized according to their codicological and palaeographic The presence of these clergymen in Sigüenza contributed to shaping the cathedral’s characteristics, we were able to determine the existence of thirty unknown codices: canonry, whose members have followed the Rule of Saint Augustine since at least 1144 eleven breviaries, ten antiphonaries, three graduals, one missal, one psalter, one

74 75 CANTUS PLANUS 2014 RUIZ TORRES - RUBIO SADIA — LITURGICAL FRAGMENTS OF THE DIOCESE OF SIGÜENZA tropary-prosary, one passionary, one mixed manuscript with chants for the mass and The decoration is not a remarkable element in our manuscripts, an aspect that confirms the divine office, plus the abovementioned bifolio of polyphony. The table below shows their clear functional nature. Nevertheless, some miniatures of value can be found in the dating of the thirty manuscripts. From it, we can see that the most intense periods in codices 1, 4, and 11. In the most ancient samples (up to the early thirteenth century), terms of copying of codices are between the twelfth and the thirteenth centuries (thirteen initials are penned in red and black, with hardly any ornamentation. The blue ink, manuscripts) and between the fifteenth and the sixteenth centuries (nine manuscripts). albeit present in codex 1, starts appearing more frequently from the thirteenth century. Both periods of activity can be understood within the historical context. During the In the same century, filigree decoration began to be used. twelfth and thirteenth centuries, Roman-Frankish traditions were still being copied and assimilated in the Iberian Peninsula (Rubio Sadia 2011, p. 329). In fact, this is the period The coordinates set for the musical notation are perfectly in tune with the Spanish in which the missal and nearly all of the breviaries in our corpus (ten of the eleven) liturgical books of the same period. Until the fifteenth century, chants are copied in were copied; this is consistent with the importance of these texts in guaranteeing the Aquitaine notation. The staff is virtually always present, with the exception of fragment complete recitation of the liturgical canon. Instead, most of the antiphonaries (seven of 1.1. In all of the manuscripts written before the fourteenth century, the reference the ten) were copied between the fifteenth and sixteenth centuries, reflecting changes in line is marked in dry point or in grey. From that century on, red-ink lines can be plainsong performance practices that were taking place at the time. The transmission found as well. The most ancient sample using square notation also dates back to the of the melodic corpus, hitherto mainly oral, was transformed into a ‘visual’ activity and fourteenth century, although, as is usual in Hispanic sources, square notation was memory was relegated to secondary status as the direct reading of books became a not predominantly used until the fifteenth century. The four-line staff is found only in crucial requirement to ensure its intonation (Ruiz Torres 2012, pp. 58-59). The pattern codex 22; all of the other square-notation samples use the five-line staff. The preference of musical notation was also undergoing changes during the fifteenth and sixteenth for the five-line staff is a usual feature in Spanish plainsong sources (Ruiz Torres 2012, centuries, with Aquitaine notation, the system hitherto in use on the Peninsula, pp. 201-202). gradually being superseded by square notation (Corbin 1952, p. 265). The state of conservation of the different fragments is rather inconsistent. The Table 1. Dating of the manuscripts readability is generally acceptable, apart from a few exemplars whose writing surface is quite worn. Handling of the exemplars over time is obvious in the form of cut-out or ripped parchment, the presence of marginal notes and writing attempts, as well as flyleaves resulting from the binding of some fragments.

Two of the manuscripts in the collection are particularly interesting and worthy of musical analysis: an antiphonary dated around 1100 and the polyphonic manuscript from the fourteenth century. The antiphonary contains repertoire destined for the celebrations of James the Apostle, Saint Nazarius, and the Beheading of John the Baptist. The first feast is of exceptional interest, because it demonstrates the existence of a proper office of Saint James in the Iberian Peninsula before the Codex Calixtinus was written. We might even speak specifically about a pre-Calixtinus office. But there is an even more important fact: the discovery of several textual and melodic concordances between the Sigüenza manuscript and the Codex Calixtinus. On the one hand, there 11th-12th 12th 12th-13th 13th-14th 14th 14th-15th 15th 15th-16th 1 1 13 3 1 1 1 9 is the responsory Alme perpetue, with just a textual correspondence and, on the other, the antiphon Honorabilem eximii, with text and music concordances. Curiously, the composition Alme perpetue appears in the Calixtinus as antiphon, while in the Sigüenza antiphonary it figures as a responsory. Further details about this music have been As far as palaeography is concerned, the types of script used and their evolution provided in two recent articles (Ruiz Torres 2015, pp. 395-417; 2017, pp. 79-94). through time are faithful to the Iberian traditions. The most ancient samples (codices 1 and 2) are copied in Carolingian script, albeit with quite sharp lines. The remaining For analysis of the polyphonic manuscript, we turned to David Catalunya for assistance. codices use Gothic script, with rather squared letters. The cursive features increase The first preliminary results of our research were presented at the Medieval and within the late production.

76 77 CANTUS PLANUS 2014 RUIZ TORRES - RUBIO SADIA — LITURGICAL FRAGMENTS OF THE DIOCESE OF SIGÜENZA

Renaissance Music Conference, held in Certaldo in 2013. Two more recent articles Appendix: Index of the analyzed liturgical fragments In this appendix we offer a brief relationship of the liturgical fragments analysed herein, grouped by codices about this fragment were published in Studi Musicali (Catalunya 2014, pp. 41-82; Ruiz and sorted chronologically for each of the archives examined. Torres 2014, pp. 83-90). The manuscript we are discussing here includes four four- voice pieces in Franconian notation consisting of one unidentified piece, a fragmented Archive of the Sigüenza Cathedral (Cathedral collection) Ave Maria in folio A, and two amens in folio B. The most extraordinary fact discovered No. 1 Antiphonarium, eleventh century ex. - twelfth century in. by Catalunya is that the first of the Amens is quoted in the treatise De musica mensurata No. 1.1 (Jacobus apostolus / Nazarius) No. 1.2 (Decollatio Joannis Baptistæ) (1279), best known as the Anonymous of Saint Emmeram. Specifically, the treatise No. 2 Antiphonarium-graduale, twelfth in. (Hebdomada iii post Pascha / Dominica iv post Pascha / transcribes a brief two-voice example, called ‘Ave Maria hoquetato’, with the purpose In Ascensione Domini / Hebdomada iii Quadragesimæ) of illustrating the hocket technique. Thanks to the finding of this manuscript, we No. 3 Breviarium, twelfth century med. - thirteenth century in. know that this short piece was taken from a truly colossal four-part hocketed Amen No. 3.1 (Dominica ii Quadragesimæ) belonging to an enormous conductus cum caudiis. The origin of the manuscript is still No. 3.2 (Decollatio Joannis Baptistæ / Felix presbyter et Adauctus) not totally clear. Certain codicological aspects (the use of a rastrum for the ruling of the No. 4 Missale, twelfth century ex. - thirteenth century in. (Feriæ post Pentecosten / Stephanus) staves) and notational features (the presence of punctum divisionis), point us to foreign No. 5 Graduale, twelfth century ex. - thirteenth century in. No. 5.1 (Dominica ii Adventus) scriptoria, although some details in the decoration seem to prove that the manuscript No. 5.2 (Feriæ v et vi ante Nativitatem Domini) was copied in Spain. Examination of the cathedral’s documentation suggests that it No. 6 Troparium-Prosarium, twelfth century ex. - thirteenth century in. (Sanctus tropes) may also have been used in Sigüenza. A local diploma dating from 1343 is crucial in No. 7 Breviarium, twelfth century ex. - thirteenth century med. that regard. In it, it is determined that there must be a master to teach the choirboys No. 7.1 (Stephanus) and other priests both chant and polyphony. No. 7.2 (Feria v in Cœna Domini / Feria vi in Parasceve) No. 7.3 (In Ascensione Domini / Feriæ ii et iii post Ascensionem)

To conclude, we believe that this work is necessary to the in-depth study of the liturgical No. 8 Breviarium, twelfth century ex. - thirteenth century med. (Polycarpus / Agnes virgo / Conversio Pauli) and musical life of the diocese of Sigüenza during the Middle Ages. Furthermore, No. 9 Breviarium, twelfth century ex. - thirteenth century med. (De Trinitate) it allows us to immerse ourselves in aspects that go beyond its church, such as the No. 10 Breviarium, twelfth century ex. - thirteenth century med. (Dominica iv Adventus) abovementioned concordances found with the Codex Calixtinus and the Anonymous of Saint Emmeram. This confirms how apparently marginal sources, like the membra No. 11 Breviarium, twelfth century ex. - thirteenth century med. (Dominica de Passione) disiecta, actually have great value. The indisputable value of these sources in Spain No. 12 Breviarium, twelfth century ex. - thirteenth century med. (Feriæ Hebd. i Quadragesimæ) should encourage the development of studies fostering its localization and description. No. 13 Breviarium, thirteenth century ex. - fourteenth century in. (Feriæ iv et v de Job) No. 14 Fragment with polyphonic music, thirteenth century ex. - fourteenth century in. No. 15 Antiphonarium, fourteenth century (Martinus, episcopus Turonensis / Cæcilia virgo) No. 16 Antiphonarium, fifteenth century (Dominicus, fundator Ordinis Fratrum Prædicatorum) No. 17 Antiphonarium, fifteenth century ex. - sixteenth century med. No. 17.1 (Stephanus) No. 17.2 (Thomas, episcopus Cantuariensis / In Traslatione Jacobi) No. 17.3 (In Dedicatione Ecclesiæ / Augustinus, episcopus Hipponensis) No. 17.4 (Petrus apostolus) No. 17.5 (Petrus apostolus) No. 17.6 (Dominicus, fundator Ordinis Fratrum Prædicatorum) No. 17.7 (Dominicus, fundator Ordinis Fratrum Prædicatorum) No. 17.8 (Michael archangelus) No. 18 Antiphonarium, fifteenth century ex. - sixteenth century med. (Augustinus, episcopus Hipponensis / Decollatio Joannis Baptistæ) No. 19 Antiphonarium, fifteenth century ex. - sixteenth century med. (Antoninus) No. 20 Antiphonarium, fifteenth century ex. - sixteenth century med. (In Traslatione Jacobi) No. 21 Psalterium, fifteenth century ex. - sixteenth century med. (Dominica per annum?) No. 22 Graduale, fifteenth century ex. - sixteenth century med. (In Dedicatione Ecclesiæ?)

78 79 CANTUS PLANUS 2014 RUIZ TORRES - RUBIO SADIA — LITURGICAL FRAGMENTS OF THE DIOCESE OF SIGÜENZA

Archive of the Sigüenza Cathedral (Medinaceli collection) Bibliography No. 23 Breviarium, twelfth century med. – thirteenth century in. (Hebdomada iv post Pascha) Blázquez Adrián (1988), El señorío episcopal de Sigüenza: economía y sociedad (1123-1805), Guadalajara: No. 24 Breviarium, fourteenth century ex. - fifteenth century (Petrus et Paulus) Institución Provincial de Cultura.

Catalunya David (2014), ‘Medieval Polyphony in the Cathedral of Sigüenza: A New Identification of a Parish house of San Juan Bautista – Atienza Musical Example quoted in the Anonymous Treatise of St Emmeram (1279)’, Studi Musicali n. s. 5/1, pp. 41-82. No. 25 Graduale, thirteenth century med. – fourteenth century in. (Commune unius martyris)

Corbin Solange (1952), Essai sur la musique religieuse portugaise au Moyen Âge (1100-1385), Paris: Les Provincial Historic Archive of Guadalajara belles lettres (Collection portugaise 8).

No. 26 Breviarium, twelfth century ex. – thirteenth century in. (Dominica i Quadragesimæ) Fletcher Richard A. (1978), The Episcopate in the Kingdom of León in the Twelfth Century, Oxford: Oxford University Press. No. 27 Antiphonarium, twelfth century ex. – thirteenth century in. No. 27.1 (Suffragia sanctorum) García Ángeles ed. (2004), Testigos de la historia de Sigüenza: Historiografía seguntina a través de las fuentes No. 27.2 (Suffragia sanctorum) documentales, Madrid: Librería Rayuela. No. 28 Antiphonarium, fifteenth century ex. - sixteenth century med. No. 28.1 (Commune unius confessoris) García Miguel A. - Pérez José M. eds. (2009), Enciclopedia del Románico en Castilla-La Mancha: No. 28.2 (Commune virginum) Guadalajara, 2 vols., Aguilar de Campoo: Fundación Santa María la Real. No. 29 Passionarium, fifteenth century ex. - sixteenth century med. Hernández Francisco J. (1996), Los cartularios de Toledo. Catálogo documental, Madrid: Fundación Ramón No. 30 Antiphonarium, fifteenth century ex. - sixteenth century med. (In Ascensione Domini) Areces (Monumenta Ecclesiæ Toletanæ Historica, Series I: Regesta et Inventaria Historica 1).

Herrera Antonio (1985), ‘La Marca Media de Al-Andalus en tierras de Guadalajara’, Wad-Al-Hayara 12, pp. 9-26.

Layna Serrano Francisco (1945), Historia de la villa de Atienza, Madrid: Csic

Mansilla Demetrio (1994), Geografía eclesiástica de España. Estudio histórico-geográfico de las diócesis, 2 vols., Roma: Instituto Español de Estudios Eclesiásticos.

Martínez Juan A. (1985), ‘El obispo don Bernardo de Agen y Santa Librada, virgen y mártir’, Anales Seguntinos 1-2, pp. 87-100.

Martínez Pilar (1990), Urbanismo medieval y renacentista en la provincia de Guadalajara: Sigüenza, un ejemplo singular, Ph. D. dissertation, Madrid: Universidad Complutense.

Minguella y Arnedo Toribio (1910-1913), Historia de la diócesis de Sigüenza y de sus obispos, 3 vols., Madrid: Tip. de la Revista de Archivos, Bibliotecas y Museos, 1: Desde los conienzos de la diócesis hasta fines del siglo XIII.

Muñoz M. del Carmen (1987), La Catedral de Sigüenza (Las fábricas románica y gótica), Guadalajara: Cabildo de Sigüenza.

Olea Álvarez Pedro (2009-2011), Sigüenza entre las dos Castillas y Aragón. Historia social, política y religiosa de las tierras de su obispado hasta 1300, 2 vols., Bilbao: Printhaus; vol. 1: 2009; vol. 2: 2011.

– (2014), Sigüenza entre las dos Castillas y Aragón. Historia social, política y religiosa de las tierras de su obispado hasta 1500 III, 2 tomes, Bilbao: Printhaus.

Reilly Bernard F. (1982), The Kingdom of León-Castilla under Queen Urraca, 1109-1126, Princeton: Princeton University Press.

Rubio Sadia Juan Pablo (2011), La recepción del rito francorromano en Castilla (ss. XI-XII), Città del Vaticano: Libreria Editrice Vaticana (Monumenta Studia Instrumenta Liturgica 61).

80 81 CANTUS PLANUS 2014

Rebekka Sandmeier Ruiz Torres Santiago (2012), La monodia litúrgica entre los siglos XV y XIX. Tradición, transmisión y praxis musical a través del estudio de los libros de coro de la catedral de Segovia, Ph. D. dissertation, Madrid: Universidad Complutense (http://eprints.ucm.es/22332/1/T34688.pdf) Imposing European culture on the Cape Colony: – (2014), ‘Reconstructing the Past: The Documentary Context of the Sigüenza Ars Antiqua Fragment’, medieval manuscripts in the Grey collection Studi Musicali n. s. 5/1, pp. 83-90.

– (2015), ‘¿Vestigios del corpus viejo-hispánico en la composición ibérica de canto llano? El oficio pre- Calixtino de Santiago apóstol’, Revista de Musicología 38/2, pp. 395-417.

– (2017), ‘New Evidence concerning the Origin of the Monophonic Chants in the Codex Calixtinus’, In 1861 George Grey, governor of the Cape Colony, donated his private collection of Plainsong and Medieval Music 26/2, pp. 79-94. books and manuscripts to the South African Public Library (Grey 1861). The collection

Ruiz Torres Santiago - Rubio Sadia Juan Pablo (2017), ‘Catálogo de los fragmentos litúrgico-musicales del can be divided into two very distinct sections: on the one hand there are books on, and obispado de Sigüenza (siglos XI-XVI)’, Medievalia 20/1, pp. 225-264. in African languages, all of which date from the nineteenth century. On the other hand

Ubieto Antonio (1989), Listas episcopales medievales, Zaragoza: Anubar. there are items of European origin. These are mostly early printed books, incunabula and manuscripts, some of which contain music notation. The manuscripts in the Ximenez de Rada Rodericus (1987), Historia de rebus Hispanie sive Historia Gothica, ed. Juan Fernández collection date mainly from the thirteenth to the fifteenth century. Many of them are Valverde, Turnhout: Brepols (Corpus Christianorum. Continuatio Mediævalis 72). illustrated and most of them are liturgical. Most of the music items in the collection are plainchant and found in liturgical books, with some fragments of musical notation in the bindings of other books. Only one manuscript contains polyphonic music.

This article focuses on the European section of the Grey Collection and in particular on the manuscripts related to the church of Santa Cecilia in Trastevere, Rome. In 2014 a major research project on these manuscripts was initiated at the South African College of Music, University of Cape Town, since many medieval manuscripts from that church are today in the Grey Collection of the National Library in Cape Town.

As shown in table 1, twelve extant manuscripts have been associated with the church of Santa Cecilia in Trastevere, either as having been written and/or used at the church in the Middle Ages. Four of them are today found in the Grey Collection.

Table 1. Manuscripts associated with the church of Santa Cecilia in Trastevere, Rome

Type Library and shelf-mark Time

Haimo of Auxerre: GB-Ob Ms Add. D, 104 (28719) 1069 Commentaries on Pauline Epistles Gradual CH-CObodmer Ms 74 1071 Bible V-CVbav Ms Vat. Barberini lat. 587 1097 Epistolary US-NHub Ms 1000 11th century Bible New York, Brooklyn Museum 11th century Homiliary ZA-Csa Ms Grey 6b8 11th century Legendary ZA-Csa Ms Grey 48b4–5; 11th or 12th early seventeenth century copy: century V-CVbav Ms Vat. lat. 6075-6076

82 83 CANTUS PLANUS 2014 SANDMEIER — MEDIEVAL MANUSCRIPTS IN THE GREY COLLECTION

in Quaritch’s catalogue of 15 April 1859 and reached Cape Town sometime late in Bible V-CVbav Vat. lat. 12958 12th century 1859 or early in 1860. Quaritch himself had bought the manuscripts at an auction of Passional V-CVbav Vat. lat. 5736 12th century Sotheby & Wilkinson in late March to early April 1859, where they were advertised in Passional V-CVbav Vat. lat. 5696 12th century the catalogue of 5 April as belonging to «a Consignment from Abroad added by kind permission of M. Libri» (Casson 1959-1960, p. 15). At least three of the manuscripts Psalter and hymnal ZA-Csa Ms Grey 6b4 13th century (ZA-Csa Mss Grey 6b2, 6b4 and 6b8), however, did not come into the Sotheby & Psalter and hymnal ZA-Csa Ms Grey 6b2 14th century Wilkinson auction directly from Guglielmo Libri, but as part of the unsold stock from another London book dealer, John Thomas Payne, who had closed his shop in 1850 and moved to Rome. Payne had probably bought the manuscripts from Libri when the Of the Santa Cecilia manuscripts in the Grey Collection in Cape Town, the homiliary latter moved to London in 1848. Even though Libri was a professor of mathematical contains no music notation. It was written during the eleventh century – possibly physics by profession, he was an avid collector of rare books and manuscripts. Already at the church of Santa Cecilia, and it was certainly used there (Stroux 1985). The early in his life he had gathered a collection of about 40,000 volumes. In 1830 he had legendary, which also contains no music notation, was written for the church during to flee Italy due to his political involvement in the revolution. He took the collection the late eleventh or early twelfth century (Stroux 1990). The Vatican Library houses with him to Paris, and – besides teaching at the Sorbonne and becoming a member of an early seventeenth-century copy of this manuscript (V-CVbav Vat. lat. 6075-6076), the academy of sciences – he started to deal in rare books and manuscripts. Because of which forms a companion volume to gradual from the church of Santa Cecilia in the his knowledge of medieval manuscripts the French government soon commissioned Martin Bodmer Collection (CH-CObodmer Ms 74; Lütolf 1987). The two composite him to compile a catalogue of manuscripts in the libraries of the Departements. While office books were written around 1300 and used at the church (Bezuidenhout 1990a, he was doing this, he was accused of having stolen some of the manuscripts he was 1990b, 1991). Both contain music notation. supposed to catalogue. When the matter was brought before a court in 1848, he fled to London taking with him approximately 30,000 volumes of books and manuscripts. 1. How did the Santa Cecilia manuscripts get to Cape Town? Since he arrived penniless in London, he started to auction his collection (Giacardi Refer to table 2 to trace the journey of the Santa Cecilia manuscripts into the Grey 2005, Stroux n.d. b). With Libri, the trail of the Santa Cecilia manuscripts, which ended Collection (Casson 1959-19/60, Stroux n.d. b, Steyn 2002). up in the Grey collection, gets cold. How Libri acquired the manuscripts can only be speculated on: he possibly did so while he was still in Italy and perhaps even legally. Table 2. Journey of the Santa Cecilia manuscripts into the Grey Collection The manuscripts may have left the church of Santa Cecilia in Trastevere only in the nineteenth century. They may have been cleared out in conjunction with building Owner Place Time Ms Grey works at the church and monastery in 1823 – but all of this is speculation.

Church of Santa Cecilia Rome 11th to 14th century 6b2, 6b4, 6b8, 48b4-5 As an aside, consider the prices of these manuscripts as stated in the catalogues by unknown unknown 15th to 19th century 6b2, 6b4, 6b8, 48b4-5 Sotheby & Wilkinson and Quaritch (Casson 1959-1960, pp. 31-33). Table 3 shows, how within the space of ten days the price of the manuscripts increased by thirty to fifty Guglielmo Libri Pisa - Paris - London ca. 1823-1850 6b2, 6b4, 6b8, 48b4-5 percent. John Thomas Payne London 1848? 6b2, 6b4, 6b8 Table 3. Prices of the Santa Cecilia manuscripts Sotheby & Wilkinson London 5 April 1859 6b2, 6b4, 6b8, 48b4-5

Bernard Quaritch London 15 April 1859 6b2, 6b4, 6b8, 48b4-5 Ms Grey Sotheby & Wilkinson (5 April 1859) Quaritch (15 April 1859)

George Grey Cape Town 1859/1860 6b2, 6b4, 6b8, 48b4-5 6b2 £ 8 £ 10.10s

6b4 £ 10 £ 15

Grey purchased them from Bernard Quaritch – a London book dealer, whose 6b6-8 £ 29 £ 36 catalogues he received and from whom he bought books and manuscripts regularly 48b4-5 £ 30 £ 42 (Casson 1959-1960, p. 12; Kerr 2006, pp. 147-151). The manuscripts were advertised

84 85 CANTUS PLANUS 2014 SANDMEIER — MEDIEVAL MANUSCRIPTS IN THE GREY COLLECTION

Even though the acquisition of most items in the European section of the Grey 7a28 southern Germany 1523 square Collection in Cape Town took place in a slightly haphazard way – like the purchase of the Santa Cecilia manuscripts – it is known what type of items Grey was seeking to add Versiculary 6b12 Belgium 16th century square to his collection. Concerning the acquisition of early printed books, Grey, for example, Laudarium 3b12 Italy 16th century White mensural tells the antiquarian booksellers T. & W. Boone that he is looking for English imprints, especially Caxton, for any early English poets, for any other black letter books, and for first prints – not necessarily English – of classical authors (letter dated 28 August 1857, Kerr 2006, p. 261). However, in relation to the Medieval and Renaissance manuscripts Grey apparently strove to acquire two manuscripts of each type (processionals, it is only possible to guess that Grey was looking for similar items as in his Auckland antiphoners, graduals etc), aiming for different regions or times in these two copies. collection, which, unlike the Cape Town collection, shows a more systematic Stroux sees this as evidence of the purpose of the collection: for didactic reasons, in order acquisition of manuscripts. In the Auckland collection Grey collected mainly hymnals to show the development of writing and book-making, Grey collected a great variety of and psalters, belles lettres and bibles (Kerr 2006, pp. 205-250). The collection in Cape manuscripts covering diverse notations, manuscript types, regions and times (Stroux n.d. Town reflects Grey’s aims in acquisition only to a certain extent: due to the slow postal a, p. 1). The only exception to this apparent aim for diversity is the cluster of manuscripts service and tough competition among book collectors, Grey could get very little of associated with the church of Santa Cecilia in Trastevere, Rome (Stroux 1985). what he was looking for in relation to early printed books and manuscripts (Short- The following part of the article discusses why these European liturgical manuscripts Title Catalogue 1977). In addition, the musical items did not form a priority in Grey’s were included in the Grey collection and why the collection was left in Cape Town, acquisition and collection strategy.1 However, in relation to manuscripts containing contextualising the collection within the thoughts and ideas on culture and civilisation music notation a pattern emerges when looking at all liturgical manuscripts in the current in nineteenth-century Britain. In a further step the article will consider the collection, as can be seen in table 4.2 purpose of the collection as it changed over time from a private collection to a public Table 4. Music notation in the manuscripts of the Grey Collection (Stroux n.d. a) library.

2. Why did Grey bring the Santa Cecilia manuscripts to Cape Town? Type Ms Grey Provenance Time Notation Even though Douglas Varley calls the Grey Collection a «mirror of Western culture» Evangeliary 4c1 Greece 12th century ekphonetic (Varley 1968), this statement is not true, when the collection is compared to current cultural trends in Europe at the time of its compilation. Its content certainly does not 6b3 Germany or Netherlands 1520 square and Gothic reflect European culture in the middle of the nineteenth century, it ratherreflects Gradual 7a27 Roggenburg or Augsburg 13th century square and Metz European thoughts and ideas about culture and civilisation at that time. The collection 48b3 Italy 15th or 16th square relates to several developments in nineteenth-century Britain: firstly humanitarianism century and a drive for education for all, and secondly the view of religion, specifically Psalter and hymnal 6b4 central Italian (Rieti) 13th century central Italian Christianity, as the basis of civilisation.

6b2 central Italy 14th century square The general views among the political and intellectual classes in Britain during the Antiphoner 4b5 Münster 13th century gothic nineteenth century continued to combine a conservative Christian outlook with liberal humanitarian attitudes. This combination produced the view that all human races 4c7 Dijon 15th century square are equal, albeit at different stages of development (see Cultural Anthropology). In 3c23 France 16th century square the Cape Colony this view led to highly egalitarian society in legal terms, even more so than under apartheid rule. All citizen in the Cape Colony had equal rights, and any Processional 2a17 southern Germany 16th century square male who owned property and/or earned an income was allowed to elect the Cape Colony’s parliament. Thus voting rights were not based on race (Cape Province). But here, of course, lies the catch: in order to own or rent property (a very European 1. The same holds true for the Africana section in the collection, which to the author’s knowledge contains just one hymnbook (Xhosa) with music notation. concept in any case) money is needed. Therefore the route to equity lies in working

2. Fragments and music in early printed books are not included in the table. to earn money. The knowledge to perform this work, in turn, is imparted through

86 87 CANTUS PLANUS 2014 SANDMEIER — MEDIEVAL MANUSCRIPTS IN THE GREY COLLECTION education, which lies at the hands of missionaries and is dispersed only in conjunction the chosen possession of men. […] It [the art of writing] related, with a wondrous new contiguity with the conversion to Christianity. Through the conversion to Christianity, education, and perpetual closeness, the Past and Distant with the Present in time and place; all times and all and work, people at the lower stages of civilisation could then eventually reach a places with this our actual Here and Now (Carlyle 1841). higher stage. This idea of ‘cultural evolution’ was developed in the wake of theories Grey bought the complete works of Carlyle for his library in 1866 and echoes Carlyle’s on biological evolution (see Cultural Evolution). It is based not so much on Darwin’s words in his manuscript ‘Old Books’: model as later theories of cultural evolution are, but rather on unilinear models like that of Jean-Baptiste de Lamarck. Early anthropologists like Lewis H. Morgan put forward Dear old book that hast for nigh four centuries presented these and such like words of comfort, theories of cultural evolution in the second half of the nineteenth century, arguing to so many men, in so many lands, and that now in South Africa as I sit lone upon this Christmas that education and religion are mankind’s means of progression from savagery via morn dost recall to my mind in this far land, that the strong planes of yon rocky mountain are the barbarism to civilisation. In Ancient Society (1877) he claims: Lord’s […] Faithful old volume that hast for nigh four centuries, never ceased thus to speak divine truths, in City, and in desert, in Palaces, which from thy gorgeousness thou must have over time As it is undeniable that portions of the human family have existed in a state of savagery, other dwelt in, or in the lonely mountain side where thou now talkest to me (Kerr 2006, p. 257). portions in a state of barbarism, and still other portions in a state of civilization, it seems equally so that these three distinct conditions are connected with each other in a natural as well as necessary Grey apparently strove to be a hero in the sense of Carlyle, both as a man of letters and sequence of progress (Cultural Anthropology, 2014). as a politician. During his time as governor of the Cape Colony, Grey strove for peaceful settlements between the British colony and the Boer republics, and was recalled to The culture of the indigenous societies, seen as at a lower level of civilization, can England in 1859-1860 owing to his plans for a union between the Boer republics and therefore act as a mirror of the earlier stages of European culture, Morgan argues: the British colony. His actions towards indigenous people were ambiguous: on the one hand he promoted peace treaties between the British and the Xhosa in the Cape, Since mankind were one in origin, their career has been essentially one, running in different but uniform channels upon all continents, and very similarly in all the tribes and nations of and the British and the Zulu in Natal, on the other hand he also saw and realised mankind down to the same status of advancement. It follows that the history and experience of the opportunities to gain land for the colony. Within the colony Grey improved the American Indian tribes represent, more or less nearly, the history and experience of our own administration and infrastructure, such as roads, railways, irrigation, and hospitals, remote ancestors when in corresponding conditions (Cultural Evolution, 2014). and he encouraged general education run by missionaries to be made available to colonists (Peires 2003, pp. 74-82). When he left the Cape Colony to become governor Based on the belief that all races are one species and equal, this linear concept of of New Zealand in 1861, he donated his private collection of books and manuscripts to progression also assumes that all cultures have the same outlook and goal. Grey’s the South African Public Library (Grey 1861). speech on the twenty-sixth anniversary of the South African Library in 1855 (Kerr 2006, p. 159) reflects these concepts: Here the ideas of Morgan are reflected again. For the Xhosa within the Cape Colony all of these privileges were accessible – on one condition: they had to take on European The highest degree of civilisation may subsist with the greatest degree of vice, and of human want and misery. We should not undervalue human learning, and should gratefully accept the aids and culture and religion, to co-operate with and therefore become dependent on white rule advantages it affords us; delighting also in the many sources of mental recreation which it opens (Peires 2003, pp. 292-326). On the other hand, the Xhosa that held on to their traditions up to us, enjoying these in moderation, as we do all other good gifts. But we should never forget and beliefs, and did not surrender to European culture and tradition, that is: those that it is a spiritual education for eternity which is man’s chief duty here upon earth, and that it is section of the indigenous people who according to Morgan were still at the barbarian only from this source that permanent happiness or improvement can spring. stage of civilisation and did not want to progress – they were ruthlessly extinguished or forced to take on European ways following the ‘great Xhosa cattle killing’ (Peires The Grey Collection is also informed by Thomas Carlyle’s thoughts about the unity 2003, pp. 263-291). of mankind and the ability of every person to grow through education and work (Thornton 1983). According to Carlyle the hero, the person who leads an exemplary As a result, historians are confronted by an ambivalent picture of Grey’s personality life, need not be military or political, but can also be a ‘man of letters’, since in books and government, which has been interpreted in various ways. During colonial and nothing, no work, no thoughts, no ideas are ever lost. apartheid times he was seen as a benevolent benefactor of indigenous people (Milne In Books lies the soul of the whole Past Time; the articulate audible voice of the Past, when the 1899, pp. 123-171; Rutherford 1961, pp. 291-442). However, in the 1990s the political body and material substance of it has altogether vanished like a dream. […] All that Mankind has aspects of his governance were re-evaluated with, for example, Jeffrey Peires promoting done, thought, gained or been: it is lying as in magic preservation in the pages of Books. They are a more critical view of his government (Peires 2003).

88 89 CANTUS PLANUS 2014 SANDMEIER — MEDIEVAL MANUSCRIPTS IN THE GREY COLLECTION

3. The purpose of the collection The manuscripts related to the church of Santa Cecilia in Trastevere, Rome, are such Grey’s desire to become a hero in Carlyle’s terms, can also be seen in the purpose of the a discovery: these liturgical items, which came into Grey’s collection by coincidence collection. The European section of the collection originated in the private interest of and whose music notation was seen merely as an added extra, were only rediscovered Grey as a bibliophile, and in his plans to conduct his own research on the manuscripts in 1984 (Dolbeau 1984) and today represent some of the few extant sources on Roman in the collection: liturgical practice in the Middle Ages from one particular church. All manuscripts related to the church of Santa Cecilia in Trastevere that contain musical notation have For thirty years one of the chief delights of my life has been to collect a Library, which I hoped been investigated by plainchant scholars and some have been edited: the Gradual would form the charm and recreation of my middle life and of my old age. Thus I have gone on was published in a facsimile edition with an extensive introduction by Max Lütolf indulging dreams of a tranquil and literary life, which was to be a compensation for the constant (1987) and its music described by Thomas Connolly (1975). Joseph Dyer wrote on the activity of my early years. A part of this plan was to publish, from ancient manuscripts, new editions of works, which I regarded as being capable of great improvement (Grey 1861). lectionary tones in the Epistolary (Dyer 2004). Christoph Stroux (n.d. a; 1985; 1990) investigated the manuscripts in the Grey Collection that contain music notation and Here Carlyle’s thought, that no ideas get lost in history, surfaces again. In addition, was especially interested in the Santa Cecilia connection whereas his student Morne Grey reasons that man has custodianship of these ideas for future generations. For Bezuidenhout looked at the two composite office books ZA-Csa Mss Grey 6b2 and 6b4 this reason the purpose of the collection changed over time. It started as a private in particular (Bezuidenhout 1990a, 1990b, 1991). However, even the two manuscripts collection and this aspect is especially visible in the European section, which shows in the Grey Collection that contain no musical notation can give insight into the Grey’s personal interests, for example, in Caxton prints and early English poets. The liturgical practices at the church of Santa Cecilia in the eleventh and twelfth century. private collection of old books and manuscripts shows Grey’s aim of conserving knowledge. By donating his private collection to the South African Public Library the Today the Grey Collection at the National Library of South Africa, Cape Town, presents purpose of conservation changes to one of research and education. The collection is a difficult inheritance. The ideas underlying the collection – evolutionism and Carlyle’s seen as an investment into the future of the Cape Colony. ideas – formed the legitimation of British colonialism and apartheid. Both saw the different races as being at various developmental stages of civilisation, and argued that What I have laid up for myself I can neither use nor enjoy, yet it is selfishly shut up from other the more civilised nations must assume guardianship for the indigenous people. In men, who might profitably use it and greatly enjoy it. I had hoped that, after my death, this Library, 1949, for example, the Federal Council of the Nederduitse Gereformeerde Kerk issued being left to some new country, might prove to it a treasure of great value, to some extend helping a statement on separate development: to form the mind of each of its generations, as they came following on. […] I believe South Africa will be a great country, that Cape Town, or its vicinity, will, for many reasons, be the point of chief [The aim of apartheid is not…] oppression but development, growth, upliftment, more privileges education for its young men (Grey 1861). and rights according to one’s own abilities, talent and potential. […] Whether we like it or not we are guardians of the coloureds and the natives too and we shall have the right to give reckoning to The donation therefore had pragmatic reasons: it was to aid the furtherance of God about our guardianship (Kinghorn 1997, p. 145). knowledge for the inhabitants of the Cape Colony as a public library, the basis for education and possibly a university. Even though the value of the individual books How to deal with the European inheritance and the underlying ideas of colonial and manuscripts or the collection as a whole may not have been immediately visible, collections in a postcolonial context, is currently investigated by the Archive and Public it formed a reservoir for future research discovering ideas inherent in the collection. Culture Research Initiative based at the University of Cape Town.3 In his address to the members of the New Zealand Society on 26 September 1851, Grey states:

Nor let us forget that knowledge is progressive; that the next age may be more learned than that which pre-ceded it; and that observed facts which, in the present state of human knowledge, cannot be so combined with others, as to prove of any great or striking utility, may, even within a few years from this time, be found of very great value and worth, for that generation more learned, more wise than ourselves, for whose advent we are now preparing the material and intellectual world (Kerr 2006, p. 155). 3. The Archive and Public Culture Research Initiative (www.apc.uct.ac.za) is a loose assembly of researchers and postgraduate students from diverse disciplines, who are working on the different aspects of colonial and postcolonial archives. Two musicologists (Morné Bezuidenhout and the author) and a scholar of English Literature (Hedley Twidle) are looking at the content and position of the Grey Collection in today’s South Africa.

90 91 CANTUS PLANUS 2014 SANDMEIER — MEDIEVAL MANUSCRIPTS IN THE GREY COLLECTION

Bibliography Milne James (1899), The Romance of a Pro-Consul being the Personal Life and Memoirs of the Right Hon. Sir George Grey, K.C.B., London: Chatto & Windus. Bezuidenhout Morné P. (1990a), An Italian Office Book of the Late Thirteenth Century, Cape Town: South African Library. Peires Jeffrey B. (2003), The Dead Will Arise: Nongqawuse and the Great Xhosa Cattle-Killing of 1856-7, Jeppestown: Jonathan Ball. – (1990b), Two Unknown Sources of Roman Chant in the Late Middle Ages: The Manuscripts Grey 6b2 and 6b4 in Angelo Pompilio - Donatella Restani - Lorenzo Bianconi - Franco Alberto Gallo, Rutherford James (1961), Sir George Grey, K.C.B., 1812–1898: A Study in Colonial Government, London: eds., Trasmissione e recezione delle forme di cultura musicale. Atti del XIV Congresso della Società Cassell. Internazionale di Musicologia (Bologna 27 agosto - 1 settembre 1987), 3 vols, Torino: Edt, 2, pp. 77-91. Short-Title Catalogue of Early Printed Books in South African Libraries 1470–1550 (1977), Cape Town: South – (1991), New Roman Chant in the Grey Collection of the South African Library, Manuscript Grey 6b2, African Library. typescript, Degree Doctor of musicology, University of South Africa. Steyn Carol ed. (2002), Medieval and Renaissance Manuscripts in the Grey Collection of the National Library Cape Province (2015) in Encyclopaedia Britannica Online Academic Edition of South Africa, Cape Town, 2 vols, Salzburg: Institut für Anglistik und Amerikanistik. www.britannica.com (23 February 2015). Stroux Christoph (n.d. a), ‘Music in the Manuscripts of the Grey Collection of the South African Library, Carlyle Thomas (1841), On Heroes and Hero Worship and the Heroic in History. Six Lectures, London: Cape Town’, typescript, ZA-Csa, s. s. James Fraser, Lecture v: The Hero as Man of Letters. Johnson, Rousseau, Burns. https://ebooks.adelaide.edu.au/c/carlyle/thomas/on_heroes/chapter5.html (19 September 2013). – (n.d. b), ‘The Story of Guglielmo Libri’, typescript, ZA-Csa, s. s.

Casson Leslie Frank (1959-1960), ‘The Medieval Manuscripts of the Grey Collection in Saleroom and – (1985), ‘Santa Cecilia’s Books at the Cape of Good Hope: A Preliminary Report’, Ars Nova 17, pp. 51-63. Bookshop’, Quarterly Bulletin of the South African Library 14, pp. 3-33. – (1990), ‘The Legendary of Santa Cecilia in Trastevere and its Background’, paper presented at the Conolly Thomas (1975), ‘The «Graduale» of S. Cecilia in Trastevere and the Old Roman Tradition’, Journal annual conference of the South African Association for Patristic and Byzantine Studies, Johannesburg, of the American Musicological Society 28, pp. 413-458. November 1990, manuscript, ZA-Csa, s. s.

Cultural Anthropology (2014) in Encyclopaedia Britannica Online Academic Edition Thornton Robert (1983), ‘The Elusive Unity of Sir G. Grey’s Library’, African Studies 42, pp. 79-89. www.britannica.com (2 July 2015). Varley Douglas H. (1968), The Grey Collection: Mirror of Western Culture in Cornelis Pama ed., The South Cultural Evolution (2014) in Encyclopaedia Britannica Online Academic Edition African Library 1818–1868, Cape Town: Balkema, pp. 35-40. www.britannica.com/ (2 July 2015).

Dolbeau François (1984), ‘Le légendier de Sainte-Cécile-au-Trastévère retrouvé au Cap’, Analecta Bollandiana 102, p. 302.

Dyer Joseph (2004), ‘The Eleventh-Century Epistolary of Santa Cecilia in Trastevere’, Archiv für Liturgie­ wissenschaft 46, pp. 311-350.

Giacardi Livia (2005), Guglielmo Libri in Dizionario Biografico degli Italiani, vols. 89-, Roma: Istituto della Enciclopedia Italiana, vol. 65. www.treccani.it (22 February 2015).

Grey George (1861), [‘Letter to Egidius Benedictus Watermeyer’], The Cape of Good Hope Government Gazette, December 24.

Kerr Donald Jackson (2006), Amassing Treasures for All Times. Sir George Grey, Colonial Bookman and Collector, Dunedin: Otago University Press.

Kinghorn Johann (1997), Modernization and Apartheid: The Afrikaner Churches in Richard Elphick - Rodney Davenport eds., Christianity on South Africa: A Political, Social and Cultural History, Cape Town: David Philips, pp. 135-154.

Lewis Henry Morgan (2014) in Encyclopaedia Britannica Online Academic Edition, www.britannica.com (2 July 2015).

Lütolf Max (1987), Das Graduale von Santa Cecilia in Trastevere, 2 vols, Cologny-Geneve: Fondation Martin Bodmer.

92 93 James Borders

A northern Italian intermediary between Avignon and Rome? Oxford, Bodleian Library, MS. Canonici Liturgical 375 and the chants of the 1485 Pontificale Romanum

Pontificals are books of rubrics, prayers, liturgical formulas, and chants for services over which bishops preside. These include clerical ordinations, the consecration of virgins, the dedication of a church, and the consecration of an altar, among others. Such collections were already being made in the ninth century, when prayers for non-eucharistic services culled from sacramentaries were copied into libelli together with related ordines and other liturgical texts (Vogel 1986, pp. 226-230; Rasmussen 1998, pp. 503-511; Palazzo 1998, pp. 195-201). Pontificals reached the fullest extent of their medieval development when the canonist, papal attaché, liturgist, and bishop, Guillaume Durandus, created such a book for his diocese of Mende in the Languedoc between 1293 and 1295.

As would befit a man with his legal education Durandus organized his collection systematically into three distinct books broadly encompassing services involving persons, places / objects, and occasions of special significance, respectively (Vogel 1986, pp. 253-255; Andrieu 1940b). Besides being logically arranged, the Durandus pontifical was considerably more extensive than its precursors. As regards music, Durandus manuscript sources typically contain nearly one hundred notated chants, including over eighty antiphons, more than a dozen responsories, and an assortment of mass chants. By contrast, witnesses of the important Pontifical of the Roman Curia preserve at most fifty-three chants, nearly all of them antiphons and responsories.

Although Durandus arguably represents an advance over other types of pontificals with respect to organization and scope, later medieval bishops were apparently at liberty to choose the forms of pontifical services they would conduct (only after the Council of Trent did Pope Clement viii impose an authorized edition of the Pontificale Romanum in 1596). Different types of pontificals continued to be copied well into the fifteenth century, this variety being largely a matter of local tradition (Leroquais 1937; Brückmann 1973; Kay 2007). In Rome, the Pontifical of the Curia predominated; it had been redacted in the early years of Pope Innocent iii’s reign (1198-1216) and was revised over the next half century (Vogel 1986, p. 252; Andrieu 1940a, pp. 309-315; 1940b, pp. 311-312). Some of its rubrics refer to the city’s distinctive geography and mention the pope as regular celebrant (Andrieu 1940a, pp. 248-249). In fourteenth- century Avignon the Pontifical of the Roman Curia, which the papal entourage brought to France, encountered the Durandus pontifical first used in Mende some two hundred

95 CANTUS PLANUS 2014 BORDERS — A NORTHERN ITALIAN INTERMEDIARY BETWEEN AVIGNON AND ROME? kilometers to the northwest. It remains unclear which of the two types the popes of source known to me. I should hasten to observe that the ‘non-traditional’ melodies this period preferred; abbreviated listings of pontificals in papal library inventories in Ob 375 are not note-for-note matches for the 1485 editio princeps, although some are simply too imprecise (Jullien de Pommerol and Monfrin 1991). Deluxe Durandus passages are the same or similar. Yet the evidence of the readings is strong enough to pontificals were however produced in greater numbers than manuscripts of the suggest that the music editor of the 1485 print treated chant in ways reminiscent of a Pontifical of the Roman Curia in fourteenth-century Avignon (Mansari 2006). music scribe like the one who worked on the Oxford manuscript. I will thus advance a case that the ‘non-traditional’ chants did not reach Rome directly from Avignon, but The balance shifted further once the papacy returned to Italy and the Durandus indirectly by way of northeast Italy. pontifical was chosen as the basis for printed editions of the Pontificale Romanum, the first of which was published in Rome in 1485 by the German printer, Stephan Ob 375 may only be broadly dated to the fifteenth century; manuscript evidence Planck. The editors were Agostino Patrizzi-Piccolomini, master of ceremonies under permits no greater precision (Kay 2007, p. 112; Brückmann 1973, p. 450). As regards Sixtus iv, and Johannes Burckhard, Patrizzi’s successor (Vogel 1986, pp. 255-256). provenance however I have determined that the book was intended for the diocese These men’s connections to the papacy, not to mention the dedication of the editio of Ravenna based on an oath taken by abbots subject to episcopal authority on fol. princeps to Innocent viii, give the impression of official approval. The chronological 61v: «[…] Vis s[an]c[t]e eccl[es]ie ravenati v[e]l tali mich[i]q[ue] ei[us]dem ep[isco] proximity of the consecration of the new altar for the renovated Sistine Chapel in 1483 po meisq[ue] successioribus fidem subiectionem[ …] Volo […]». The manuscript and the pontifical’s publication two years later, moreover, suggests a practical link to comprises 243 numbered parchment leaves measuring 35 × 25 cm. The text script – the papacy: one or both of the masters of ceremonies, Piccolomini and Burckhard, an angular Italian rotunda – was copied in black ink with red rubrics, in two columns very likely planned the service, which figures prominently in the Durandus pontifical 96.6 cm wide, with a maximum of twenty-six lines of text; the writing area measures and includes a large number of chants. 20 × 15 cm. Large initials were decorated with floral designs in blue, green, and red on fields of gold leaf; smaller capitals are done in red or blue with vinestems in the Attracting musicological attention to the 1485 editio princeps is the fact that about one alternating color. The chant melodies are copied in black square notation on four third of its chants differ in their texts and/or melodies from those found in earlier widely spaced, red staff lines. The music scribe’s tendency to write neumes in a small Durandus pontifical manuscripts. Although clef mistakes and faulty text underlay hand relative to the wide staves renders challenging the precise determination of mar the 1485 and Planck’s second edition of 1497, I have previously maintained that pitches here and there. these items were not merely ‘place-holders’ (Borders 2012 and 2013). The early date of these books in the history of music printing, together with the connections to the Although Ob 375 contains fewer notated melodies than many other Durandus papacy, have inclined me to give the printer the benefit of the doubt. Moreover, the pontifical manuscripts, the chant texts are generally copied in full. Especially useful ‘non-traditional’ chants were copied from the editio princeps into a handful of deluxe for comparison are three chants which in the 1485 print either replace those in most manuscripts and also reprinted in a series of editions published in the first quarter of Durandus manuscripts or were added to the base text (Andrieu 1940b). These are the sixteenth century, further suggesting that the 1485 chants were actually sung. given in table 1, p. 98. All three chants are found in Ob 375; two also appear in I-Tn E.III.8, a Durandus pontifical from the second half of the fifteenth-century, altered in Rather than examining issues of performance or music-editorial problems, I wish to the 1470s to conform to Roman, and arguably papal, customs (Kay 2007, p. 185). The consider here how the ‘non-traditional’ chants, which are generally new or newly owner of the latter book, Spanish cardinal Pedro Ferriz, Roman resident and a familiar adapted compositions or type melodies, found their way into the 1485 Pontificale of the Venetian Pope Paul ii (Pietro Barbo), had a keen interest in papal ceremonies Romanum. Neither Patrizi nor Burckhard was known to have been a musician, and judging from this pontifical and a RomanCeremoniale he also commissioned (US- thorough study of extant pontifical manuscripts indicates that the ‘non-traditional’ Cn f72.1), itself related to the 1485 Pontificale. Referring to the table, the communion chants were not Roman products. Writing about the adoption of the Durandus pontifical Redime me Deus Israel substituted for the more widely transmitted Notas mihi fecisti. as the model for the 1485 print, historian Leon Strieder (2005, p. 33) characterized it The antiphon Zachaee festinans and the alleluia v Bene fundata est augment the usual as «something Roman go[ing] to Avignon and then something French return[ing] to corpus of chants in Durandus manuscripts. Rome». I propose that the situation was more complicated, at least as far as the chants are concerned. Our main focus will be GB-Ob 375 (GB-Ob MS. Canonici Liturgical 375, henceforth Ob 375) a fifteenth-century Durandus pontifical from northeastern Italy, the chants in which related more closely to those in the 1485 print than any other

96 97 CANTUS PLANUS 2014 BORDERS — A NORTHERN ITALIAN INTERMEDIARY BETWEEN AVIGNON AND ROME?

Table 1. Three chants substituted into or augmenting the 1485 Pontificale Romanum Table 3 surveys the three remaining 1485 chants with text variants. Ob 375 again with manuscript antecedents (sources of the Durandus Pontifical) agrees in all instances with the editio princeps. Two variants – debet in the antiphon

Liturgical Ambulate sancti Dei ingredimini and eorum in the antiphon Via sanctorum – are function Incipit Feast Substitute for Sources found solely in this manuscript and the 1485 and later prints. The remaining Durandus pontifical sources listed in table 3 were intended for southern France and Communio Redime me Deus Israel De consecratione Communio GB-Ob Can. Lit. 375 virginum Notas mihi fecisti text only southern Italy and date from the fourteenth or fifteenth centuries. In summary, the evidence of added or substituted chants as well as text variants points to a connection I-Tn E.III.8 between a northeastern Italian manuscript source – among surviving Durandus text only witnesses documented in Ob 375 – and the 1485 editio princeps. Antiphona Zachaee festinans De ecclesie GB-Ob Can. Lit. 375 (Cao 5515) dedicatione (alia antiphona) Table 3. Three antiphons with text variants in in the 1485 Pontificale Romanum with manuscript (addition) antecedents (sources of the Durandus Pontifical) I-Tn E.III.8 (alia antiphona)

Alleluia v Bene fundata est De ecclesie GB-Ob Can. Lit. 375 Ambulate sancti Dei GB-Ob Can. Lit. 375 ac cimiterii text only ingredimini in civitatem domini edificata est enim vobis reconciliatione ecclesia nova ubi populus adorare debet (addition) maiestatem domini

(no Cao; debet] debeat […] in most Durandus manuscripts)

Besides similarities in the selection of chants, five antiphons in the 1485 print share text Pax eterna ab eterno huic domui GB-Ob Can. Lit. 375 variants with Ob 375 and a few other manuscripts in its milieu. The texts of two involve pax perennis verbum patris […] F-Pn lat. 733 F-Psg 143 added words, which are more useful as regards filiation than deletions or substitutions. (Cao 4252: ab eterno patre huic domui […] See table 2, in which the variants are underlined. In the first, Ancilla Christi for the in most Durandus manuscripts) consecration of virgins, the added word habere survives in four manuscripts besides Ob 375 (I have already mentioned I-Tn E.III.8; GB-Ob Canonici Liturgical 362 was copied in fifteenth-century Venice. The provenances of the two other fifteenth-century Italian Via sanctorum facta est recta y y GB-Ob Can. Lit. 375 pontificals listed in table 2 have yet to be determined). In the case of the second antiphon, et iter eorum z preparatum est. F-C97 Signum salutis for the laying of a new church’s cornerstone, the words domine Ihesu F-Pn lat. 733 (cf. Cao 5374, 2534[b]: V-CVbav Reg. lat. 1930 Christe are found in Durandus pontificals solely in Ob 375 and the 1485 and later prints. facta est recta] recta facta est; z GB-Ob Can. Lit. 375 eorum] sanctorum […] Table 2. Texts added to antiphons in the 1485 Pontificale Romanum with manuscript antecedents in most Durandus manuscripts) (sources of the Durandus Pontifical)

Ancilla Christi sum GB-Ob Can. Lit. 375 ideo me ostendo I-Tn E.III.8; To gain a sense of the ‘non-traditional’ chant melodies in Ob 375 and how they compare servilem habere personam GB-Ob Can. Lit. 362 both with earlier Durandus manuscripts and the 1485 print, we will examine on two (Cao 1393) I-Tn I.IV.1 (+ I-Rn 1464, PRC-Durandus hybrid, olim Rome, antiphons with text variants that lack concordances in office antiphoners (noC ao). Holy Cross of Jerusalem Abbey) In example 1 (p. 100), reading 1 of Signum salutis (cf. table 2) comes from F-C Ms. 97, a mid-fourteenth-century Durandus pontifical intended for Arles, a setting quite Signum salutis pone domine Ihesu Christe GB-Ob Can. Lit. 375 in loco isto similar to those found in most other Durandus manuscript sources. This melody et non permittas introire may be assigned to the fifth mode with cadences on the supertonic at the midpoint angelum percutientem and on the final at the end. The ambitus expands in the second of the two phrases, (no Cao) the highpoint being d.

98 99 CANTUS PLANUS 2014 BORDERS — A NORTHERN ITALIAN INTERMEDIARY BETWEEN AVIGNON AND ROME?

Example 1. Antiphon Signum salutis (no Cao) The reading on line 2, from Ob 375, employs a first mode type melody. Cadences at the 1: F-C Ms. 97 (Pontifical of Guillaume Durandus), fol. 76 midpoint and end are on the tonic, D. The ambitus of both the first and second phrases 2: GB-Ob. Can. Lit. 375 (Pontifical of Guillaume Durandus), fol. 88 3: Pontificale Romanum, Rome: Stephan Planck, 1485, fol. 120v of this melody is essentially the same. The texture is simpler than the version of the 4: Pontificale Romanum, Venice: Lucantonio Giunta, 1520, p. 115 chant in most Durandus pontificals (line 1) and, for that matter, the reading from the 1485 print (line 3). Note that the contours of the 1485 melody as well as the version in the 1520 Pontificale Romanum (line 4) published in Venice by Giunta generally follow that of Ob 375, particularly the opening phrase.

The 1485 melody might be construed as third mode, but I have reservations about taking this and other versions in the print at face value. Apparently so did the anonymous music editor of the 1520 Venetian Pontificale Romanum (line 4). In this and later sixteenth-century editions the second phrase was transposed down a major second relative to the 1485 version. It may have been that a first mode melody similar to that in Ob 375 figured as printer’s copy for the 1485 editio princeps and the type- setter botched it. A reading like that in Ob may also have afforded guidance to the 1520 Venetian music editor, who corrected the longstanding mistake.

Example 2, p. 102 compares three settings of Ambulate sancti Dei ingredimini, an antiphon for the relics procession connected with the consecration of a church (cf. table 3).

Line 1 is from F-AIXm 13, a Durandus pontifical copied in Avignon in the second quarter of the fourteenth century for Armand de Narcès, archbishop of Aix (Kay 2007, p. 1); lines 2 and 3 are from Ob 375 and the 1485 print, respectively. Like settings in most other Durandus manuscripts, the line 1 melody is in third mode. It opens with a recitation on F, a gesture repeated to extend the final phrase at populus adorare. The texture is neumatic-to-melismatic, with important words – Dei, Domini, enim vobis, ecclesia nova, and maiestatem – set to melodic flourishes. Distinctive features of this widely transmitted version of the chant include syllabic ascents from the finalis on ingredimini and edificata.

Unlike the line 1 melody, the setting in Ob 375 (line 2) is syllabic-to-neumatic; its ambitus and cadences on G allow for a seventh-mode classification. Note in particular the final cadence on G, approached by way of the subtonium modi. The opening gesture, F-a-c, which repeats four times mainly at phrase openings, is however unexpected in a chant of this mode. A similar opening figure is found in the Gregorian repertory in fifth mode gradual verses including V.Congregate illi sanctos ejus [Gr. Ex Sion] and V. Confidentur Domino misericordiae [Gr. Misit Dominus]. Perhaps the Ob 375 scribe was aware of the formula’s structural function but unconcerned about its association to a different mode in earlier Gregorian chants.

On line 3, the ‘non-traditional’ setting in the 1485 print appears related to both the others. Note the syllabic textured ascents on ingredimini and edificata as in

100 101 CANTUS PLANUS 2014 BORDERS — A NORTHERN ITALIAN INTERMEDIARY BETWEEN AVIGNON AND ROME?

Example 2. Antiphon Ambulate sancti Dei ingredimini (no Cao) Durandus manuscripts; in 1485 only the first of these is slightly developed by a 1: F-AIXm 13 (Pontifical of Guillaume Durandus), fol. 106 melodic sequence on civitatem. The greater affinity however is with Ob 375, which 2: GB-Ob Can. Lit. 375 (Pontifical of Guillaume Durandus), fol. 107 3: Pontificale Romanum, Rome: Stephan Plannck, 1485), p. 149v the 1485 setting resembles in the closing phrase, beginning ubi populus. The two melodies share the same final cadence, which was also used in the first phrase of the 1485 melody on the first occurrence of the word domini.

These two examples scratch the surface of a far larger investigation, but I hope that the evidence of the musical readings, shared repertory, and common fund of text variants offered here is sufficient to suggest some connection between the milieu of Ob 375, copied for Ravenna, and the 1485 Pontificale Romanum,published in Rome. With this possibility in mind, let us return to the initial question: how might the ‘non-traditional’ chants have found their way to the papal circle and ultimately to a Roman printer’s shop in the late fifteenth century? To address this, let us begin by considering the nature of the transmission of later medieval pontificals and their chants.

It may have been that Bishop Durandus intended his pontifical solely for the diocese of Mende. Within a few decades of his death in 1296 however copies were being made in southern France, where thirteen of twenty-two extant fourteenth-century sources originated. At first, then, the spread of Durandus pontifical might appear to have followed a pattern of regional transmission comparable to that of earlier pontificals, but I would argue that the particular conditions of book production at the time played a new and important role. The first step in appreciating this history is recognizing the emergence of professional scribal workshops as producers of deluxe liturgical and chant books in the later Middle Ages.

Art historical research has demonstrated that more than half the known fourteenth- century southern French Durandus pontificals were copied and/or decorated in Avignon (Mansari 2006, p. 162). The wealthy churchmen who commissioned these deluxe books need not have resided there or in nearby episcopal sees – such was the case with F-AIXm 13, noted above. The professionalization of scribal work in the later Middle Ages has not inclined scholars of liturgy to doubt the authority of the texts in these pontifical – Andrieu relied on them for his edition – but the practical and conceptual distances separating scribes in professional ateliers and print shop workers from earlier cantors and singer-scribes involved in musical performance should give us pause: in the eleventh through the thirteenth centuries, pontificals were copied in monastic or chapter scriptoria where scribes worked practically within earshot of their churches, a situation quite different from fourteenth-century luxury book production in places like Avignon.

It may have been that the ranks of later professional copyists included clerics, but copying pontificals must always have been challenging work due to the infrequency

102 103 CANTUS PLANUS 2014 BORDERS — A NORTHERN ITALIAN INTERMEDIARY BETWEEN AVIGNON AND ROME? with which some episcopal services were performed as well as the multiplicity of type were made and illuminated in the very same ateliers as Durandus pontificals. exemplars. The editors of the 1485 editio princeps, Patrizzi and Burckhard, made While it may at first seem counterintuitive, the continued vitality of the Pontifical a point of complaining about the latter in their preface. Moreover, with respect to of the Roman Curia outside Rome in the later fourteenth century owed to Avignon’s the chants in particular, one should bear in mind a further challenge: twenty-two prestige despite the need to adapt the rubrics to suit local conditions. antiphons and three responsories in Durandus were sung exclusively in pontifical services. Thus about a quarter of the antiphons lack concordances in antiphoners The fifteenth century witnessed a decline of the desirability of Pontifical of the and other readily available chant books. These chants may have been unfamiliar Roman Curia; only seven manuscripts, one illuminated in Avignon, survive from this even to singers, let alone professional scribes who may not have been clerics. At the time. Meanwhile, the dissemination of the Durandus pontifical expanded to include same time, late medieval scribes were growing increasingly accustomed to taking more dioceses, prominent among them those in northern Italy. The preponderance liberties with melismas even in stably transmitted melodies, which are here and of evidence suggests that one such manuscript – a lost fifteenth-century Durandus there abbreviated or altered in Durandus pontificals, including those copied in pontifical, copied in northeast Italy, and probably similar in many respects to Ob Avignon. Professionals music scribes’ lack of familiarity with less widely recognized 375 – figured in the production of the 1485 editio princeps. The exemplar may have pontifical chants may have afforded them increased license to make changes or even come from the Veneto, considering how deftly the music editor of the 1520 Guinta substitute a known melody or melody type for an unknown one. I suspect this is Pontificale Romanum corrected mistakes that had been reprinted in six prior what the scribe of Ob 375 did. editions. Concordances between Ob 375 and I-Tn E.III.8, copied for a bishop who was a familiar of a Venetian pope, reinforce the possibility of Venetian provenance. The second step to understanding the difference between the earlier and later As far as music in late medieval and early modern pontificals is concerned, the ways medieval transmission of chants in pontificals is recognizing the importance of in which certain chant melodies were handled in Rome after the Schism was not certain prestigious ecclesiastical centers in the later Middle Ages, a period of papal simply a matter of something French – the Durandus pontifical – being sent there, monarchy. The wide dissemination of Durandus pontificals in the fourteenth and but the importation of workshop-based production methods as well as books and fifteenth centuries beyond southern France into northern and central France, northern perhaps singers from northeast Italy. Italy, and Spain had to do with the preferences of the bishops who commissioned them, but very little is known concerning for certain about the provenances of these sources, let alone the workshops that created them. This is in part because they contain so few regional features, if any at all. Indeed, important Durandus witnesses for Amiens and Autun, at once geographically distant from Avignon but close enough philologically to the lost archetype for Andrieu to have collated readings from them in his edition, suggest a set of considerations for copying that differed drastically from those that reflected liturgical customs and chant traditions of contiguous regions in eleventh through the thirteenth centuries. Earlier scribes, who may have been cantors themselves, adapted the texts in pontificals to suit their recollections of the actual episcopal liturgies; later professional scribes did not do this because they not only lacked the experience but also because they worked under the shadow of church authority, especially that of the papacy. Thus late medieval patterns of manuscript dissemination speak not just to the concentration of skilled labor in a few ecclesiastical centers but also to the prestige invested in places like Avignon and its projection across great distances aided by the professionalization of scribal work.

This pattern began to emerge in Rome as early as the thirteenth century when the popes typically resided there. By the fourteenth century, when manuscripts of the Pontifical of the Roman Curia were being copied in fairly large numbers, most sources were copied in Rome (Andrieu 1940a, pp. 3-224). Later in Avignon, pontificals of this

104 105 CANTUS PLANUS 2014

Jurij Snoj Bibliography

Andrieu Michel (1940a), Le Pontifical romain au Moyen Âge, 3 tomes, Città del Vaticano: Biblioteca The antiphoner of Izola Apostolica Vaticana, tome 2: Le Pontifical de la Curie Romaine au XIIIe siècle (Studi e testi 87).

– (1940b), Le Pontifical Romain au Moyen Âge,3 tomes, Città del Vaticano: Biblioteca Apostolica Vaticana, tome 3: Le Pontifical de Guillaume Durand (Studi e testi 88).

Borders James (2012), A (Brief) Introduction to Post-Tridentine Pontifical Chants in Cantus Planus Study Group of the International Musicological Society: Papers read at the 16th meeting, Vienna, Austria, 2011. Vienna: Hollinek, pp. 58-65. Izola/Isola is a small Slovenian town on the northern shores of the Istrian peninsula. It

– (2013), The Transmission of Non-Cao Antiphons in Later Medieval Pontificals in Cantus Planus Study belonged to the Republic of Venice until its collapse at the beginning of the nineteenth th Group of the International Musicological Society: Papers read at the 15 meeting, Dobogók , Hungary, century. From the ecclesiastical point of view, Izola was made part of the Diocese of 2009, 3 vols., Ottawa: The Institute of Mediaeval Music, 2, pp. 423-56. Koper/Capodistria, situated just a mile away from Izola. Said diocese, believed to have Brückmann John (1973), ‘Latin Manuscript Pontificals and Benedictionals in England and Wales’, Traditio been established already in the sixth century by saint Nazarius of Koper, functioned as 29, pp. 391-458. a suffragan diocese of the Patriarchate of Aquileia. Jullien de Pommerol Marie-Henriette - Monfrin, Jacques (1991), La Bibliothèque pontificale à Avignon et à Peñiscola pendant le Grand Schisme d’Occident et sa dispersion: inventaires et concordances, 2 vols., Rome: École française de Rome (Collection de l’École française de Rome 141). The main church of Izola was the church of Saint Maurus of Pore /Parenzo. It was (and still is) the town’s parish church where there was also a chapter of canons. Its Kay Richard (2007), Pontificalia: A Repertory of Latin Manuscript Pontificals and Benedictionals, Lawrence: University of Kansas. history is not very well known. In his extensive description of the Diocese of Koper from the end of the seventeenth century, Bishop Paolo Naldini established that the Leroquais Victor (1937), Les pontificaux manuscrits des bibliothèques publiques de France, 3 vols., Paris: [Protat]. chapter of Izola was founded sometime in the fourteenth or early fifteenth century; as Mansari Francesca (2006), La Miniatura ad Avignone al Tempo dei Papi (1310-1410), Modena: Panini. he gathered from the documents (which he only quoted in general terms), there was very likely no chapter in Izola around 1308, but there was certainly one in 1421, when Palazzo Eric (1998), A History of Liturgical Books from the Beginning to the Thirteenth Century, transl. Madeleine Beaumont, Collegeville, MN: The Liturgical Press. Gieremia Pola, Bishop of Koper, appointed a certain Francesco d’Antonio a canon in Izola. The chapter was not a large one; the number of canon posts seems to have Rasmussen Niels Krogh (1998), Les pontificaux du haut moyen âge: Genèse du livre de l’évéqu, Louvain: Spicilegium sacrum Lovaniense (Études et documents 49). varied through centuries, and in Naldini’s time there were only four canons in Izola (Naldini 1700, pp. 335-341). Strieder Leon (2005), The Promise of Obedience: A Ritual History. Collegeville, MN: The Liturgical Press.

Vogel Cyrille (1986), Medieval Liturgy: An Introduction to the Sources, trans. and revised by William Storey The only remaining witness to the liturgical practices in the chapter of Izola is an and Niels Rasmussen, Washington, DC: The Pastoral Press. antiphoner which is still kept in the church archives. The antiphoner of Izola consists of four large volumes. It is written in square notation and, given its appearance, may be dated to the fifteenth century. As usually, the first volume includes the temporale from Advent to Easter, and the second one the temporale from Easter to the end of the liturgical year; the third volume is the sanctorale; and the fourth one is the common of saints. The four volumes appear to have been conceived and compiled as a complete whole; the only digression from the conventional order of offices is the last fascicle of the commune sanctorum volume, which contains the office of saint Maurus of Pore /Parenzo, i.e., the patron saint of the church, and the office of the feast of Corpus Christi. The fact that this fascicle is included as part of the commune sanctorum volume, even though it contains the office of a particular saint, seems to indicate that it was attached to the manuscript at a later date, sometime after the completion of the antiphoner as a whole.1

1. Centuries-long usage caused substantial damage to the volumes: there are many lacunae and several later additions; some folios have been removed from individual volumes and inserted erroneously into the commune sanctorum volume.

106 107 CANTUS PLANUS 2014 SNOJ — THE ANTIPHONER OF IZOLA

A little-known manuscript such as the antiphoner of Izola raises basic questions. Firstly, feasts (e.g., Saint Francis, Saint Clara, Francis’ Stigmata). Moreover, some important its liturgical order should be specified and its liturgical peculiarities or special items, ubiquitous feasts are missing as well (e.g., Nicolaus, Gregory, James the Apostle). The if any, identified. Secondly, the historical environment in which the manuscript was comparison of the sanctorale of Izola and Haymo’s Ordo breviarii reveals an interesting compiled should be determined by means of a more thorough analytical inspection fact: the antiphoner of Izola contains all of the feasts of Haymo’s Ordo breviarii for and an appropriate appraisal of its notation. which Haymo prescribes proper chants, but none of those for which Haymo’s breviary prescribes common chants or just prayers and lessons. In the antiphoner of Izola, In order to assess the liturgical features of the manuscript, let us inspect some of its Nicolaus is therefore missing, by virtue of the fact that the feast includes no proper characteristic segments. The list of responsories for Advent Sunday in the antiphoner chants in Haymo’s breviary, but only those from the common of saints. There are only of Izola matches a number of other manuscripts, including Franciscan antiphoners, two exceptions to this principle: 1) the antiphoner of Izola includes the feast of Mary’s antiphoners presenting the Roman tradition, monastic antiphoners following the Visitation, which was introduced only after Haymo, and 2) there are no Franciscan Subiaco-Melk reform and others, but no antiphoners from Aquileia or Cividale, even feasts in the antiphoner of Izola (as mentioned above). though Izola belonged to the Patriarchate of Aquileia.2

The comparisons of the series of antiphons sung to Benedictus and Magnificat on Table 1. Sanctorale of the Antiphoner of Izola Sundays after Pentecost provide more accurate results.3 Manuscripts that include the same series are less numerous and more specific: they are Franciscan, except for the Feast Folio Feast Folio antiphoner from Saint Ulrich and Afra in Augsburg.4 This monastery followed the reform of Melk, and adopted the rite of the monastery of Subiaco, which was, in fact, Andreae 002 In Sanctorum Ioannis et Pauli 081 the monastic version of the Roman rite, which, in turn, was originally Franciscan Luciae 010 Petri et Pauli 085 (Klugseder 2013, pp. 178-186). In Vigilia S. Thomae Apostoli 013 In Commemoratione S. Pauli 091 According to these comparisons, the antiphoner of Izola may either belong to the In Vigilia S. Agnetis Virginis 013 Visitationis 100v liturgical order of the Franciscans or represent the Roman rite, which was, after the revision of Haymo of Faversham in the 1240s, basically the same as that of the In Conversione S. Pauli 021v Mariae Magdalenae 113 Franciscans (Hiley 1993, pp. 594-595). Indeed, no matter what part of the temporale In Festo S. Agnetis Secundo 029 In Vincula S. Petri 120 of the antiphoner of Izola is chosen, it agrees almost completely with Haymo’s Ordo breviarii (van Dijk 1963, pp. 17-195), be it Sundays in Lent, Holy Saturday, or summer In Purificatione 029 Laurentii 121 historiae. In Vigilia S. Agathae 037v In Vigilia Assumptionis 132

5 Let us turn to the sanctorale of the manuscript (table 1). As can be seen from the In Cathedra S. Petri 046v In Decollatione Ioannis 142 table, the sanctorale is very basic and does not include any name characteristic either of the region or of the ecclesiastical province. There are no Aquileian saints (e.g., In Annuntiatione S. Mariae Virginis 051v In Nativitate Beatae Mariae 145 Hermagoras and Fortunatus, Cantius, Cantianus, Cantianilla and Protus, Hellarus and In Festo S. Marci et Aliorum Apost. 057 In Exaltatione S. Crucis 148 Tatianus), no office of Saint Nazarius of Koper/Capodistria, but also no Franciscan Philippi et Iacobi 058 Michaelis 151

2. This comparison makes use of the search facility ‘Responsories for Advent’, available on the website Cantus Planus In Inventione S. Crucis 061 In Vigilia Omnium Sanctorum 158v Regensburg. The series in the Antiphoner of Izola agrees with twenty-nine of the sources listed therein.

3. The comparison makes use of the search facility ‘Post-Pentecost Cantica Antiphons’, developed by David Hiley and Ioannis ante Portam Latinam 063 Martini 167 available on the website Cantus Planus Regensburg.

4. Three of the manuscripts included in the aforementioned search facility have exactly the same series: H-Bu lat. 119 In Inventione S. Michaelis 063 Caeciliae 175v (a Franciscan antiphoner, fourteenth century); I-Ac 694 (a Franciscan antiphoner, thirteenth century); and I-Ac 693 (a Franciscan antiphoner, thirteenth century). Three other manuscripts are very close to the series of the antiphoner of Izola: In Vigilia S. Ioannis 071 Clementis 183 D-Mbs Clm 4306 (Augsburg, Saint Ulrich and Afra, 1501); D-Ma 12o Cmm 1 (a Franciscan antiphoner, thirteenth century), and CH-Fco 2 (a Franciscan antiphoner, thirteenth century).

5. Table 1 shows just names in modern spelling; octaves and commemorations have been omitted.

108 109 CANTUS PLANUS 2014 SNOJ — THE ANTIPHONER OF IZOLA

Not very many manuscripts with a sanctorale like this could be found. Among the Table 2. From the office of the Commemoration of Saint Paul antiphoners of the Cantus database, just a few come close to that of Izola;6 however, no historical or other connections appear to exist among them, except that some Ordo breviarii Antiphoner of Izola, fol. 100r-v exhibit the Franciscan liturgical order. Closest to the antiphoner of Izola seems to be Notandum quod per octavam Notandum est quod per octavam the antiphoner from the Franciscan monastery of Dubrovnik, whose sanctorale is ant. Gloriosi principes terre cum V. Constitues ant. Gloriosi principes terre dicitur almost identical to that of Izola. Disregarding the extent of the offices, there is only one eos principes dicitur ad Magnificat ad Magnificat. difference between both sources: the omission of the Feast of the Visitation from the et ant. Petrus apostolus Ant. Petrus apostolus et Paulus cum V. In omnem terram V. In omnem terram dicitur Dubrovnik manuscript ( izmi 2013, pp. 624-626). ad Benedictus ad Benedictus. cum oratione Deus qui hodiernam diem This examination leaves no doubt that the antiphoner of Izola follows the liturgical Ant. Gloriosi principes terre Gloriosi principes order of Haymo of Faversham, except that it does not include the Franciscan feasts. It is ant. Petrus apostolus ant. Petrus apostolus Ad vesperas ant. Iuravit dominus cum Ad vesperas ant. Iuravit dominus cum therefore a pure example of an antiphoner of the Roman rite. Manuscripts of this kind reliquis. Psalmi de apostolis. reliquis antiphonis et psalmis de apostolis. could presumably be used everywhere; in Izola, obviously, a Roman antiphoner was Pro sancto Iohanne cap. Audite insule Pro sancto Iohanne ad Magnificat used notwithstanding the fact that it was in the territory of the Aquileian ecclesiastical hymnus Ut queant laxis ant. Ingresso Zacharia et duplica ratione festi V. Fuit homo R. Cui nomen ad Magnificat commemoratione sancti Pauli province. It is also safe to assume that manuscripts of this kind (i.e., manuscripts that ant. Ingresso Zacharia Et duplicatur ratione could be used by any church lacking a particular liturgical tradition) were produced festi commemorationis sancti Pauli. Oratio regardless of the place of their destination. In this respect, they may be considered Deus qui presentem diem forerunners of the prints. Deinde fit commemoratio de apostolis: ant. et fit commemoratio de apostolis Gloriosi principes terre V. Constitues eos ant. Gloriosi principes V. Constitues eos. R. Memores oratio Deus qui hodiernam diem A detailed inspection of the relationship between Haymo’s text and the antiphoner of Izola reveals that the latter faithfully follows the wording and expressions of the former, but omits all directions concerning recited or spoken items (i.e. not sung), were to be transported to Africa, his homeland, but a strong wind carried the vessel such as prayers, lessons, and chapters. To illustrate the procedure, a portion from the across the seas to the shores of Istria. The inhabitants of Pore /Parenzo (a Croatian end of the office of the Commemoration of Saint Paul as it appears in both sources town on the western shores of the Istrian peninsula) carried the relics to the church (table 2)7 may be quoted (van Dijk 1963, p. 149, the antiphoner of Izola, sanctorale and began venerating Saint Maurus as a saint and their patron (Saints d’Istrie, pp. 371- volume, fol. 100r-v). The comparison makes quite clear that the antiphoner of Izola 380). The office of Saint Maurus of Pore seems to be unique to the antiphoner of Izola. stands among the descendants of Haymo’s Ordo breviarii and the fact that it omits Although the saint’s name appears in the calendar of Aquileia (Camilot-Oswald 1997, Franciscan feasts should be understood as a clear indication that it was deliberately pp. cxxii, cxxxv), the office cannot be found in any Aquileian or other hitherto known prepared for a secular (non-Franciscan) church. manuscript. It seems to have been limited to the area of Venetian Istria.8

Yet the antiphoner of Izola does not lack a local stamp. As already mentioned, there is In the remainder of this article, let us concentrate on the notation of the antiphoner a fascicle attached to the commune sanctorum volume which contains the offices of of Izola. It is a risky and uncertain undertaking to try and identify particular hands Saint Maurus of Pore /Parenzo (Snoj 2010) and that for the feast of Corpus Christi. Saint in late manuscripts. As the technique of writing seems to have been rather invariable Maurus of Pore , the patron saint of the church of Izola, is an obscure figure. According and uniform by the end of the Middle Ages, many different north Italian manuscripts to his legend, he was a Roman martyr from the end of the third century. His relics of the fifteenth century exhibit exactly the same way of shaping and putting down the characters of the script, and almost the same applies to the notation as well. In all four volumes of the antiphoner of Izola, the same way of putting down characters and 6. Cantus database, maintained by Jan Kolá ek and Debra Lacoste. The manuscripts in question are: A-Wn 1799** (a the same way of shaping notational signs may be discerned. Nevertheless, browsing Cistercian antiphoner from Rein, thirteenth century); H-Bu lat. 121-122 (a Franciscan antiphoner, fourteenth century); CH-Fco 2 (a Franciscan antiphoner, thirteenth century); D-Ma 12o Cmm 1 (an Italian antiphoner, thirteenth century); E-SA through the folios of the volumes, one must observe changes in the outer appearance 5-8 (fourteenth century); I-Ac 693-694 (thirteenth century); HR-Hf Cod. E (from the Franciscan monastery in Dubrovnik, ca. 1400).

7. The abbreviated words in the antiphoner of Izola are written out in table 2, except for those designating the genres. The directions concerning the commemoration of John the Baptist were obviously not understood by the scribe of the 8. A fifteenth-century breviary originating from Pore /Parenzo contains another office of the same saint (Ah 28, 70-71) antiphoner of Izola and are therefore corrupted. that bears no resemblance to that in the antiphoner of Izola.

110 111 CANTUS PLANUS 2014 SNOJ — THE ANTIPHONER OF IZOLA of the script, whether gradual or abrupt. Especially intriguing in this respect is the last three-note torculus using the oblique stroke. Yet the oblique stroke forms part of all fascicle of the commune sanctorum, which includes the office of Saint Maurus (and porrectus signs, should they consist of three or more notes. Three or more descending appears to have been attached to the volume after its completion). The first folio of the notes are always rendered by slanting diamonds, and these always follow either a office of Saint Maurus seems to have been written by a different hand from the one that virga or podatus. The scandicus has two forms: a pes plus virga or a punctum plus wrote previous parts of the book; however, after some pages, the script changes and pes. But when three ascending notes (i.e., a scandicus) are followed by a descent, becomes nearly indistinguishable from the rest of the manuscript. the combination of a podatus and climacus appears (see the last sign in the seventh row); in other words, a scandicus is normally not followed by descending diamonds. The question of whether the antiphoner of Izola should be regarded as the work of Notational nuances such as these certainly have no effect on the manner of singing, one or several scribes may be addressed by comparing its script and notation to other and a minimal effect on the musical conception of chants. Yet, at the very least, they manuscripts from the same time and region, including those of Aquileia, Cividale, represent distinctive customs of scriptoria and scribes, and as such, they should not be nearby Koper/Capodistria, or even Venice. Notwithstanding the uniformity of late overlooked either by historical or by palaeographical studies. manuscripts, it was impossible to recognize the hand or the hands of the antiphoner of Izola among the contemporary manuscripts from Aquileia and Cividale (Camilot- Table 3. Notational signs in the antiphoner of Izola Oswald 1997, facsimiles after p. cxxxviii). All of them differ from the antiphoner of Izola to a greater degree than various parts of the antiphoner of Izola among themselves. The same is true for the Venetian manuscripts (Cattin 1990, pp. 115-122, 133, 140- 1 148), and not even examination of manuscripts from the nearby Koper/Capodistria9 led to identification of the hand (or hands) that wrote the antiphoner of Izola. It must therefore be concluded, although not without caution, that the antiphoner of Izola 2 came into being as the work of one specific scriptorium or even one scribe, who also copied the fascicle with the office of Saint Maurus, although the latter appears to have 3 been attached to the already existing book somewhat later. The assumption that the fascicle containing the office in question was produced in the same scriptorium as the entire manuscript might suggest that the compilation of the four volumes took place 4 within the limited area in which the cult of Saint Maurus was observed, meaning one of the Istrian towns. One can imagine that the fascicle containing the office of Saint Maurus was prepared after it became clear that the four volumes would be used by the 5 chapter of Izola. The fact that it was attached to the commune sanctorum volume (and not to the sanctorale) may just have been for practical purposes. 6

The term square notation stands for a variety of late scripts that differ not just in the graphical or personal traits of the scriptoria and scribes, but also in the structural 7 characteristics of the notation, i.e., the way in which individual signs and their different categories were composed (Szendrei 1998, pp. 274-275). Table 3 shows the 8 structural characteristics of the notation of the antiphoner of Izola: a single note is always rendered by a virga (not by a punctum). The podatus consists of two vertically aligned squares, the clivis by two successive squares. The torculus alone is composed 9 of three squares; but when it functions as a starting point of a larger sign, it may be written using the oblique stroke (the latter standing for the second and the third 10 notes, see the last sign in the fourth row). This is quite consistent, and there is no

11 9. These manuscripts are kept in the Diocesan Archives of Koper/Capodistria.

112 113 CANTUS PLANUS 2014 SNOJ — THE ANTIPHONER OF IZOLA

It is a common assertion that in late manuscripts special and liquescent signs gradually The notation of the first temporale volume of the same antiphoner of Koper (written by disappeared. However, this general statement does not necessarily apply to any one a different scribe) is also very close to that of Izola. This volume does not contain any particular manuscript. In Aquileian manuscripts, liquescents are quite frequent, and scandicus subbipunctis, yet it features another characteristic trait: in the succession in the antiphoner of Izola there are also several special signs. Some of these are quite of a clivis and a torculus the former appears written as the oblique stroke; in Izola, frequent, and others rare. With the exception of four (shown in the last row of table this melodic turn is rendered by a porrectus and clivis. The liquescent signs do appear 3, p. 113), they only differ from common signs in their last element, which indicates in this manuscript, but very rarely. It is exactly this kind of intermingling of various their functioning as liquescents. To get an idea of their frequency, let us examine some notational traits and scribes’ habits that makes palaeographical mapping (i.e., mapping folios at the beginning of the temporale volume. The first fifteen pages contain forty- of notational types), even of a smaller historical environment, methodologically so seven special signs; forty-three of these appear in places that may require a liquescent extremely difficult. note: above nasals, liquids, clusters of consonants, and above diphthongs. By far the most frequent liquescent sign consists of a slightly inclined punctum, followed by The antiphoner of Izola is not interesting because of its exclusive characteristics; it is a virga (the first sign in the eighth row). It is not quite clear what this sign stands interesting as an example of its kind, a specimen of a late antiphoner according to the for; presumably it denotes a prolonged tone, at the end of which comes a phonetic Roman rite written in square notation exhibiting specific notational traits. Manuscripts peculiarity. It is also difficult to determine the way in which this sign differs from such as this, although not the main focus of plainchant scholarship, represent an similar signs that obviously also denote some peculiarity in rendering a single tone important segment of late medieval music history. Considered from this point of view, (the signs in the eighth row). It may be that there is no clear distinction among them, the examination of the antiphoner of Izola may well demonstrate the questions and and that they just stem from different notational traditions that somehow merged in problems of late medieval plainchant history. the scriptorium in which the antiphoner of Izola was compiled. Nevertheless, it must be stressed that the notation of the antiphoner of Izola encouraged singing that paid due attention to the pronunciation of the text.

This is just one particular specimen of square notation that can also be found in other manuscripts. Nonetheless, it is not possible to locate the antiphoner of Izola historically on the basis of the structural characteristics of its notation. In the politically Venetian and ecclesiastically Aquileian area making up the environment to which Izola belonged, other specimina or types of square notation existed as well. In addition, what renders the assessment of the notation of the antiphoner of Izola (as well as notation of other manuscripts from the same area) especially problematic is the fact that the described structural characteristics do not appear just in one clearly definable group of manuscripts. Strictly speaking, this is to say that the notation of the antiphoner of Izola cannot be conceived of as a specific type of square notation, but as a combination and amalgamation of particular notational traits and habits that, intermingled with other structural characteristics, may or may not turn up in other manuscripts from the same area.

Just to illustrate the last observations, let us compare the notation of the antiphoner of Izola to some of the manuscripts from the nearby Koper/Capodistria. The notation of the commune sanctorum volume of the antiphoner of Koper, which was compiled by a certain Nazarius de Iustinopoli (i.e., Nazarius of Koper) sometime in the fifteenth century, conforms fairly well to that of Izola; yet this manuscript contains scandici subpunctis (i.e., the melodic turn in question is not rendered by the succession of a podatus and climacus), and Nazarius obviously did not make use of liquescent signs.

114 115 CANTUS PLANUS 2014

Réka Miklós Bibliography

Ah (1898), vol. 28, Clemens Blume - Guido M. Dreves eds., Leipzig: Reisland. Der Seckauer Liber ordinarius von ca. 1595 Camilot-Oswald Raffaella (1997), Die liturgischen Musikhandschriften aus dem mittelalterlichen Patriarchat (A-Gu 1566) als letztes Dokument der mittelalterlichen Aquileia, Kassel: Bärenreiter (Mmmae, Subsidia 2, Teilband 1). Salzburger-Seckauer Liturgie und Musik Cattin Giulio (1990), Descrizione delle fonti in Giulio Cattin ed., Musica e liturgia a San Marco, 4 vols., Venezia: Fondazione Levi, vol. 1, p. 17-148.

izmi Ana (2013), The Dubrovnik Antiphoner in Barbara Haggh-Huglo - Debra Lacoste eds., Papers Der Seckauer Liber ordinarius, geschrieben zwischen 1592 und 1600 vom späteren Read at the 15th Meeting of the IMS Study Group Cantus Planus, 3 vols., Lions Bay: The Institute of mediaeval Music, vol. 2, pp. 619-628. Seckauer Domdekan Georgius Huebner ist das letzte umfassende Zeugnis der mittel- alterlichen Salzburger-Seckauer Liturgie vor der Einführung des römischen Ritus in ijk D Stephen J. P. van (1963), Sources of the modern Roman liturgy, 2 vols., Leiden: E. J. Brill (Studia et 1 documenta Franciscana 1-2), vol. 2: The Ordinals by Haymo of Faversham and Related Documents. Seckau im Jahre 1600. Er beschreibt sehr ausführlich die Mess- und Tagzeitenliturgie sowie die Besonderheiten des liturgischen Jahres wie ausgedehnte Prozessionen usw. Hiley David (1993), Western plainchant: a handbook, Oxford: Clarendon Press. (Abbildung 1, p. 118). Es finden sich auch Hinweise auf Polyphonie und Orgelspiel. Klugseder Robert (2013), ‘«Secundum rubricam Romanam»: reform liturgy of Subiaco-Melk’, De musica Diese Quelle hat in der bisherigen Forschung kaum Beachtung gefunden, stellt aber disserenda 9, pp. 175-190. ein Dokument für die Liturgie und Musik dieser Zeit von besonderem Rang dar.

Naldini Paolo (1700), Corografia ecclesiastica o’ sia descrittione della citta, e della diocesi di Giustinopoli, Themen des Vortrags sind die Entwicklungen im musikalischen Repertoire in einem Venezia: Gierolamo Albrizzi. Zeitraum von 250 Jahren: die Veränderungen, Kontinuitäten und Innovationen, mit ‘Saints d’Istrie, et de Dalmatie’ (1899), Analecta Bollandiana 18, pp. 371-380. besonderer Fokussierung auf das Messrepertoire.

Snoj Jurij (2010), The Office of S. Maurus of Porec´/Parenzoin Rastislav Adamko ed., Musica mediaeva 1. Der Liber ordinarius A-Gu 1566 liturgica, Ružomberok: Katolícka Univerzita v Rožomberku, pp. 126-141. und die parallel verwendeten Musikhandschriften Szendrei Janka (1988), ‘Perspektiven der musikalischen Paläographie in der Choralforschung’, Studia Der Liber ordinarius A-Gu 1566 dokumentiert nicht nur die Seckauer Liturgie kurz musicologica 39, pp. 267-282. vor 1600, sondern zitiert häufig die dabei verwendeten liturgischen Bücher: das ‘große Graduale’ (die heute einzig erhaltene Musikhandschrift: A-Gu 17),2 das ‘große Anti- phonar’, das ‘kleine Antiphonar’, das ‘kleine Graduale’, das Responsoriale, den ‘Codex parvum’, sowie den ‘kleinen’, den ‘schwarzen’ und den ‘roten Codex’. Außerdem gibt es Hinweise, dass sowohl das Salzburger als auch das Seckauer Brevier verwendet worden sind. Die neben den Gesängen angegebenen Zitierungen aus A-Gu 1566 las- sen auf jeden Fall zwei großformatige und zwei kleinformatige Musikhandschriften vermuten. Mit Hilfe der zahlreichen Folioangaben in A-Gu 1566 war es möglich, ei- nige virtuelle Indices dieser nicht mehr erhaltenen Musikhandschriften anzufertigen, die für die weitere Forschung nützlich sind.

2. Die Messfeiern in Seckau Nachdem ich die gesamte Handschrift in moderne Schrift übertragen und die In- dices der liturgischen Gattungen angefertigt hatte, folgten im Rahmen meiner Dis- sertation (Miklós 2016) systematische Vergleichsarbeiten. Dabei habe ich zuerst die 250 Jahre lokale Messrepertoire-Entwicklung von 1345 bis ca. 1600 untersucht.

1. «Ipse cum suo capitulo proprietatibus se abdicavit et anno [M]DC breviario Romano missalisque Romanis in et extra chorum uti caepit», Amon 1960, p. 231.

2. Die Handschrift A-Gu 17, zitiert in A-Gu 1566 als ‘Graduale summum’ wurde erstmals von Franz Karl Praßl identifiziert (1987, 1, pp. 105-117).

116 117 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Abbildung 1 Abbildung 2 Der späteste Liber ordinarius aus Seckau, Graz, Universitätsbibliothek, Sondersammlungen Graduale summum, Graz, Universitätsbibliothek, Sondersammlungen A-Gu 1566, fol. 114v A-Gu 17, fol. 188v

118 119 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Die für den Vergleich verwendeten Quellen waren folgende drei Handschriften aus der Graduale summum Verwendung fand, wurde meistens aus dem kleinen Graduale Grazer Universitätsbibliothek: (graduale minus) gesungen, das heute nicht mehr erhalten ist. Mit Hilfe zahlreicher Folioangaben, die in A-Gu 1566 vorkommen (Abbildungen 3, p. 120 und 4, p. 128), war Liber ordinarius A-Gu 756 von 1345 (Behrendt 2009) es möglich, einen partiellen virtuellen Index der liturgischen Gesänge dieser nicht 3 Graduale summum A-Gu 17, ca. 1480-1510 mehr erhaltenen Musikhandschrift anzufertigen. Liber ordinarius A-Gu 1566 ca. 1592-1600 (Miklós 2016) A-Gu 1566 gibt die folgenden drei Kategorien von Formularen für die Frühmesse Auf dem ersten Blick fällt auf, dass A-Gu 17 nur die Formulare der Hauptmessen im wieder: Repertoire enthält (Abbildung 2, p. 119).

Die Messtypen betreffend ist der Liber ordinarius A-Gu 1566 differenzierter als A-Gu 17, a) Marienmesse weil hier auch die Formulare der marianischen Frühmessen vorkommen (Abbildung 3). b) das Formular der Frühmesse an den Herrenfesten und an anderen liturgischen Anlässen im Kirchenjahr, wo die Frühmesse nicht als Marienmesse zelebriert wurde c) explizit keine Frühmesse

Die Marienmesse als Frühmesse. Wie bereits zuvor erwähnt differenziert der Liber ordinarius die Marienmesse abhängig davon, ob sie an marianischen Hochfesten oder an gewöhnlichen Tagen zu feiern ist. Die gewöhnliche Marienmesse im Kirchenjahr folgt diesem Schema:

a) Formular der im Kirchenjahr vorgeschriebenen Marienmesse (Tabelle 1):

Abbildung 3. Hinweis auf Graduale minus in A-Gu 1566, fol. 279v Tabelle 1

Aus dem Liber ordinarius A-Gu 1566 geht schon deutlich hervor, dass in der Stifts­kirche Bezeichnung Introitus Nr. Verweise Seckau um 1600 täglich mehrere Messen gefeiert wurden. Grundsätzlich unterschei- vom ersten Adventssonntag bis Rorate caeli 779 A-Gu 17 det man zwei Kategorien: die Frühmesse (maturum officium oder officium diurnum) Weihnachten sowie die Hauptmesse (officium summum). An bestimmten Tagen mit höherem Fest- von 30. Dez. bis Mariä Reinigung Vultum tuum 948 k. A. rang, bei Koinzidenz zweier Feste oder bei besonderen liturgischen Anlässen war von der Vorfastenzeit bis außer den Maturum- und Summum-Messen auch eine weitere Messe, ‘sacrum’ oder Vultum tuum 1146 Graduale minus Gründonnerstag ‘prior’ genannt, üblich. Die Kanoniker lasen außerdem ihre Privatmessen. von Osterdienstag bis zum ersten Salve sancta parens 1448 Graduale minus Die Frühmesse. Das Maturum wurde gewöhnlich nach der Laudes gefeiert. Eine Ab- Adventssonntag weichung von der Regel gab es in der Fastenzeit, wo die Frühmesse nach der Prim ze- lebriert wurde.4 Die Kanoniker feierten die Frühmesse nicht als Privatmesse, sondern als Gemeinschaftsmesse – dies bestätigen die Folioangaben der dabei verwendeten b) Formular der eigenen Messe zum jeweiligen Marienfest im Kirchenjahr (Tabelle 2): Musikhandschriften.5 Abgesehen davon, dass bei einigen besonderen Anlässen das Tabelle 2

3. Die Handschrift wurde 2014 von der Autorin indiziert. Liturgischer Anlass Introitus Nr. Verweise 4. Nr. 1256 «Singulis dominicis diebus usque ad pascha dicatur prima ante maturum officium» (A-Gu 1566; alle lateinischen Abschriften erfolgten durch die Autorin). PurificationisB mv Suscepimus 1139 Graduale minus 5. Nr. 1146 «Ad maturum officium In Vultum tuum 19. Kyrie et Sanctus dominicaliter. Si fuerit binos Kyrie et Sanctus de virginibus ad placitum. Sine Gloria tamen. Gr Dilexisti iusticiam 4. Tr Audi filia 32. Of Aue Maria uel Filiae regum 27. Annuntiationis Bmv Rorate caeli 1353 Graduale minus Co Diffusa est 4. Hoc modo a septuagesima usque ad caenam domini maturum officium cantetur exceptis duobus diebus quibus proprium habet» (A-Gu 1566).

120 121 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Es ist anzunehmen, dass das Formular der Frühmesse auch bei anderen mariani- Adalram wurde in A-Gu 1566 das Laus tibi Christe anstelle des Alleluja-Rufs mit schen Hochfesten (Assumptionis Bmv, Visitationis Bmv, Nativitatis Bmv, Praesenta- dem Vers Assumpta est Maria vorgeschrieben (Tabelle 4): tionis Bmv) dem der Hauptmesse folgte, nämlich mit dem Introitus Gaudeamus. Tabelle 4 c) Marienmesse in Bezug auf die Herrenfeste. Bei den größten Herrenfesten im Kir- Nr. 1214 IN Gaudeamus chenjahr verwendeten die Kanoniker für die Frühmesse das Formular der Haupt- messe. Am zweiten Tag aber (manchmal sogar am Festtag) feierten sie die Früh- Kyrie et Sanctus de Assumptione messe nicht selten zu Ehren Marias. Am Mittwoch der Osterwoche wurde außer GR Propter ueritatem dem Summum nur das Sacrum mit Osterformular verwendet. Die Frühmesse war bereits Marienmaturum. Ebenfalls wurde die Pfingstmontags-Frühmesse – bei der Pro Alleluia cantetur Laus tibi Christe Assumpta est Maria die Chorherren nicht das kleine Graduale, sondern das Graduale summum ver- SE Congaudent angelorum chori wendeten – feierlich im erhöhten Chor beim Dreifaltigkeitsaltar (Beata Maria in Neustifft) zelebriert. Am Dienstag nach Pfingsten ist die Frühmesse als Marienmes- OF Offerentur regi se mit dem Introitus Salve sancta parens eingetragen. Erst ab dem Mittwoch nach CO Dilexisti Pfingsten wird für die Frühmesse wieder auf das Pfingstformular zurückgegriffen. Am Festtag Corporis Christi wurde die Frühmesse zu Ehren Marias gefeiert. Die gleiche liturgische Ordnung galt auch am Oktavtag. Am Fronleichnamsfest hat man Die Frühmesse der Herrenfeste. Bei den Herrenfesten im Kirchenjahr sowie zu beim Maturum – sowohl am Festtag als auch am Oktavtag – anstelle der Herren- einigen anderen liturgischen Anlässen wurde die Frühmesse nicht als Marienmesse messe die Marienmesse gefeiert (Tabelle 3). gefeiert, sondern richtete sich nach der Hauptmesse des Tages oder der Woche. In der Weihnachtsoktav bis zum dritten Tag nach dem Fest der Unschuldigen Kinder Tabelle 3 fand die Weihnachtsmesse als Frühmesse Verwendung. Am Tag der Unschuldigen

Liturgischer Anlass Introitus Nr. Verweise Kinder erklang die Frühmesse noch mit dem weihnachtlichen Alleluja-Ruf und mit der Sequenz. Für die Hauptmesse aber vertauschte der Schreiber den Ruf: statt des Feria 2 Paschae ‘De Assumptione per totum’ 1441 k. A. Alleluia wurde das Laus tibi Christe gesungen. Die Sequenz konnte auch weggelassen Feria 3 Paschae Salve sancta parens 1448 Graduale minus werden, falls der Pädagoge (der Kantor) sie nicht singen wollte.6

Feria 2 Penthecostes Gaudeamus 1709 A-Gu 17 In der Zeit nach Pfingsten wurde das Formular der Pfingstmesse von Mittwoch nach Feria 3 Penthecostes Salve sancta parens 1710 Graduale minus Pfingsten bis zum Dreifaltigkeitssonntag verwendet.7 Wie bereits zuvor erwähnt wur- de am Dienstag nach Pfingsten die Frühmesse Introitus Salve sancta parens gefeiert. Corporis Christi ‘De Assumptione’ 1715 Graduale minus Deshalb verschob sich die Pfingstmesse auf den zweiten Platz und wurde als Sacrum In octava Corporis Christi ‘De Assumptione’ 1716 [Graduale minus] zelebriert. Am Fronleichnamsfest und dessen Oktavtag wurde das Marienmaturum nicht weggelassen, sondern die Fronleichnamsmesse als Prior genommen. d) Marianische Frühmesse an den Professtagen in Seckau. Gleichzeitig mit dem Die Frühmessen an den Herrenfesten und zu anderen liturgischen Anlässen im Fest Praesentationis Mariae wurde in Seckau der Professtag gefeiert, der als ein Kirchen­jahr, bei denen die Frühmesse nicht als Marienmesse zelebriert wurde, sind besonders feierlicher Anlass galt, bei dem die liturgische Ordnung dem Hochfest wie folgt (Tabelle 5): Mariä Himmel­fahrt folgte. Ebenfalls wurde der Faschingmontag sehr feierlich zu Ehren Marias gestaltet. Deshalb ist sowohl die Haupt- als auch die Frühmesse mit 6. Nr. 932 «Hoc tamen obseruato quod hac die Gloria et in terra Ite missa est et Te deum laudamus non dicimus nisi hoc dem Formular Introitus Gaudeamus zu finden. Die Anlage des Formulars lässt den festum in dominicam occurat. Ratio huius est quia sancti qui ante passionem domini de hac uita migrabant quamuis essent Schluss zu, dass dies vermutlich der Tag der Profess-Erneuerung war, die im Li- in requie tamen plenitudinem gaudii non habebant quia Christus morte sua nondum redemerat mundum. Gr Anima nostra. Pro alleluia Laus tibi Christe cantetur. Se Laus tibi Christe patris optimi 273 aut nullum pro ut paedagogo placet» ber ordinarius A-Gu 756 nicht aufscheint. Wahrscheinlich aus Rücksicht auf den (A-Gu 1566). folgenden Ascher­mittwoch sowie wegen des Totengedenkens an den Stiftsgründer 7. Nr. 1712 «Cum reliquis feriis usque ad festum trinitatis cantetur maturum de sancto spiritu per totum» (A-Gu 1566).

122 123 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Tabelle 5 etwas kürzere Messe ohne Graduale (bzw. zweites Alleluia) und gesungene Commu- nio. Das Sacrum wurde gewöhnlich gemeinsam bis zur Wandlung gefeiert, danach Liturgischer Anlass Introitus Nr. Verweise folgte ein Salve regina.11 Anschließend wurde nicht mehr gemeinsam im Chor gesun- Nativitatis Domini Puer natus 908 Graduale minus gen, sondern der zelebrierende Kanoniker sollte die Messe still bis zum Ende lesen, während der Chor das Salve regina sang. Im Hinblick auf die Aufführungspraxis gibt es Epiphaniae Ecce advenit 993 Graduale minus in A-Gu 1566 mehrere Hinweise, dass das Salve regina von den Schülern intoniert wur- Dominica Paschae Resurrexi 1431 Graduale minus de.12 Das Salve regina konnte auch wegfallen, wie dies eine Messe für die Bruderschaft Sebastianus zeigt. In diesem Fall wird nach der Wandlung gleich die Sext gebetet.13 In inventione Crucis Nos autem 1589 Graduale minus Manchmal schreibt der Liber ordinarius das Sacrum schon vor der Frühmesse vor wie Ascensionis Domini Viri Galilei 1656 Graduale minus z. B. am Tag des Heiligen Stephanus Prothomartyrus, an dem diese Messe dem Gründer

In die Penthecostes Spiritus Domini 1708 Graduale minus Adalram (Alram) gewidmet ist. Die außergewöhnliche Feierlichkeit dieses Anlasses bestätigt nicht nur der Hinweis, dass die Messe beim Hauptaltar zelebriert wurde, son- Feria quarta Penthecostes Spiritus Domini 1712 Graduale minus 14 usque Trinitatis dern auch die Verwendung des großen Graduale A-Gu 17. Was das Formular betrifft, verfährt das Sacrum sehr unterschiedlich, wie im Folgenden dargestellt wird: Trinitatis Benedicta sit 1713 A-Gu 17

Transfigurationis Domini In excelso throno 319 A-Gu 17 a) Messformular folgt dem der Hauptmesse (Tabelle 6):

Tabelle 6 Keine Frühmesse. Es kommt auch vor, dass der Liber ordinarius explizit keine Früh- messe vorschreibt, wie z. B. am Freitag vor Pfingsten: «Maturum officium hac die non Liturgischer Anlass Introitus Nr. Verweise dicitur» (fol. 279v). Ursache dafür könnten die Quatembertage und der Bußcharakter Barbarae Gaudeamus 787 Graduale minus der Woche vor Pfingsten sein. Nicolai Statuit ei 796 Graduale minus Die Hauptmesse. Die Summum officium genannte Hauptmesse wurde meistens nach Sebastiani et Fabiani Intret in conspectu 1073 Graduale minus den kleinen Horen oder zwischen Prozession und Non feierlich zelebriert. Zahlrei- (‘fraternitatis Sebastiani’) che Zitierungen weisen darauf hin, dass die Kanoniker dabei aus dem großen Gra- Feria tertia Paschae ‘de Resurrectione 1450 Graduale minus (‘officium textoribus’) per totum’ duale (graduale summum, A-Gu 17) gesungen haben. An den größeren Festen mit vor­gesehener Vigilmesse wurde das Summum gewöhnlich am Vorabend nach der Non Lanceae et clavorum Domini (AL1–2) Foderunt manus meas 1518 Graduale minus zelebriert.8 Die gewöhnliche Tagesmesse der Fastenzeit (missa de feria) war nach der Floriani Sancti tui 1601 A-Gu 17 Non an der Tagesordnung. Falls auf diesen Tag ein Heiligenfest fiel, wurde die Haupt- Corporis Christi Cibavit 64 A-Gu 17 messe mit dem Formular des Heiligenfestes zelebriert, die Tagesmesse aber deswegen nicht weggelassen. Auch außerhalb der Fastenzeit koinzidierten nicht selten Feste; es In octava Corporis Christi ‘De Corporis Christi’ 1716 A-Gu 17 wurde aber dafür gesorgt, dass die Messen nicht wegfielen.9 Petri et Pauli Nunc scio vere 151 k. A.

Weitere Messtypen. Das Schindelamt.10 Zwischen Maturum und Summum wurde in Seckau bei den größeren Festen sowie im Rahmen anderer spezieller liturgischer 11. Nr. 320 «Post maturum officium habemus missam priorem de sancta trinitate in choro quae cantari solet usque ad Anlässe eine weitere Messe (‘sacrum’ oder ‘prior’ genannt) gefeiert, die meistens vor eleuationem. Finita eleuatione incipietur Salue regina» (A-Gu 1566). oder nach der Prim eingefügt wurde. Für die Kanoniker bedeutete dies eine zeitlich 12. Nr. 151 «Post maturum officium habetur prior apud sancte Petrum. In Nunc scio uere. Kyrie et Sanctus de apostolis. Al Tu es Petrus. Se Petre summe Christi. Of Constitues. Finita eleuatione incipietur a pueris Salue regina» (A-Gu 1566).

8. Nr. 105 «Hoc officium post nonam dicimus» (A-Gu 1566). 13. Nr. 590 «Celebraturque patrocinium super altare sancti Sebastiani. Finita eleuatione incipietur etiam sexta» (A-Gu 1566).

9. Nr. 105 «Si in octauam corporis Christi euenerit post tertiam cantetur de corpore Christi officium deinceps post nonam 14. Nr. 906 «Hora quinta hac die habetur a domino praelato requies fundatori Alramo super altare summum cum de vigilia Ioan Baptistae» (A-Gu 1566). ministrantibus et cantatur hoc officium a canonicis et domicellis una cum scholaribus per totum ut habetur in summo graduali absque tamen responsoriis quibus alias utimur fundatori sub eleuatione. Finito hoc officio cantetur Haec est 10. Bezüglich dieses Messtypes habe ich den Ausdruck von Jungmann verwendet (1962, 1, p. 173, besonders Fußnote 22). dies pro Salue regina. Deinde subiungitur prima» (A-Gu 1566).

124 125 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Messen mit vollständigen Formularen. Abweichungen vom Schindelamt kommen Liturgischer Anlass Introitus Nr. Verweise bei Festkollisionen bzw. in einigen anderen Fällen vor, bei denen mehrere Anlässe Mariae Magdalenae Gaudeamus omnes 257 k. A. mit Eigenoffizien vorgesehen sind. Im Folgenden werden sämtliche Messformulare, die weder zum Maturum noch zum Summum gehören, in diese Kategorie eingeordnet. Augustini In medio ecclesie 400 A-Gu 17 Ein weiterer besonderer Anlass für die Seckauer Marienkirche war die relativ spät Michaelis archangeli Benedicite Domino 482 A-Gu 17 zelebrierte Marienmesse am Mittwoch der Pfingstwoche. Diese wurde im erhöhten Dürnberger-Chor mit Orgel und Gesang gefeiert.15 Manchmal wurde das ganze Mess- Wolfgangi Statuit ei 591 Graduale minus formular anstelle des gewöhnlichen Schindelamtes angegeben. Folgende Beispiele voll- Martini episcopi et confessoris Sacerdotes tui 649 Graduale minus ständiger Formulare erschienen nach der anschließenden Hauptmessen (Tabelle 9).

Elisabethae Gaudeamus 686 Graduale minus Tabelle 9

Catharinae Loquebar 730 k. A. Liturgischer Anlass Introitus Nr. Verweise Andreae apostoli Michi autem 751 A-Gu 17 Ioannis evangelistae In medio ecclesie 920 Graduale minus

Philippi (1AL) Exclamaverunt 1579 Graduale minus b) Formular folgt dem des Maturum (Tabelle 7):

Tabelle 7 Diese Messen wurden nicht immer unmittelbar vor oder nach der Prim gefeiert. Bei den folgenden Beispielen liturgischer Fest- bzw. Anlasskoinzidenz sind weitere voll- Liturgischer Anlass Introitus Nr. Verweise ständige Messformulare ohne Salve regina zu finden (Tabelle 10): In octava Omnium Sanctorum Gaudeamus 632 Graduale minus Tabelle 10 Feria tertia Penthecostes Spiritus Domini 1711 Graduale minus Liturgischer Anlass Introitus Zeiltiche und örtliche Nr. Verweise Anpassung c) Formular ist weder mit dem des Summum noch mit dem des Maturum identisch Feria tertia maioris Cibavit ‘post sextam in capella 1391 Graduale (Tabelle 8): hebdomadae Lichtenstainern’ minus

Feria sexta hebdomadae Humiliavit se Iesus ‘post tertiam in sacrario’ 1372 Graduale Tabelle 8 passionis minus

Liturgischer Anlass Introitus Nr. Verweise Feria sexta quatuor Si iniquitates ‘infra processionem in 861 Graduale temporum templum parochiale’ minus Dominicis diebus infra Pascha Protexisti me Deus 1530 Graduale minus (Temp. Adv.) et Penthecosten (‘Missa fraternitatis Sebastiani’)

Mariae ad nives ‘de assumptione’ 312 k. A. Einen weiteren Messtyp, der ebenfalls weder Maturum noch Summum ist, stellt die Transfiguratio – sacrum de trinitate ‘de trinitate in choro’ 320 k. A. Tagesmesse der Fastenzeit dar. Diese Messen wurden an den Festtagen der Heiligen nach der Non gefeiert (Tabelle 11): feria sexta Penthecostes Si iniquitates 28 k. A. (+CO, absque GR)

15. Nr. 21 «Deinde subiungitur nona. Post nonam sermo habetur. Finito sermone habetur officium apud altare beatae Mariae in Neustifft de assumptione cum cantu et organo» (A-Gu 1566).

126 127 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Tabelle 11 Auf das hintere Spiegelblatt des Graduale A-Gu 17 trug er mit schwarzer Tinte zwei Messformulare nach, welche jedoch im Liber ordinarius fehlen. Hierbei sind zwei, Liturgischer Anlass Bemerkung Nr. Verweise auch in A-Gu 1566 nachweisbare Schriftarten von ihm zu erkennen (Abbildung 6).

Annuntiationis Bmv ‘Hac die post nonam habetur 1355 k. A. etiam de feria officium’

Mathiae apostolis ‘De festo missa festiue agitur ad tertiam. 1269 k. A. De feria uero ad nonam officium celebratur’

3. Das Verhältnis zwischen A-Gu 1566 und dem Graduale summum A-Gu 17 Wie bereits zuvor erwähnt hat der Schreiber in A-Gu 1566 die Folii des Graduale A-Gu 17 angegeben (Abbildung 4). Diese großformatige diastematische Musikhandschrift Abbildung 6. A-Gu 17, fol. 394 wurde unter Propst Johannes Dürnberger (1480-1510) angefertigt und noch um 1600 verwendet. Während meiner Untersuchungen wurde ich auf eine besondere Schrift aufmerksam, die nur in den Nachträgen Seckauer Quellen auffindbar ist. Das Kyrie und das Gloria mit dem Tropus Spiritus et alme orphanorum auf fol. 234r-v des Graduale A-Gu 17 verlangt weitere Untersuchungen. Das Schriftbild hat eine Neigung nach links, die sogar im Noten­bild erkennbar ist. Weiterhin fällt auf, dass der Schreiber den gotischen Schrift- typus wegen Platzmangels nicht bis zum Schluss verwenden konnte (Abbildungen 7 und 8). Abbildung 4. Folioangabe des Graduale magnum im Liber ordinarius A-Gu 1566, fol. 36

Huebners charakteristische Schrift konnte ich in anderen Seckauer Quellen nachwei- sen, z. B. auch im o.g. Graduale A-Gu 17. Huebner hat hier einige Rubriken sowie die Nummern der Folien bei den einzelnen Alleluia-Incipits nachgetragen Er verwendete dabei zwei verschiedene Nuancen von roter Tinte (Abbildung 5).

Abbildung 7. A-Gu 17, fol. 234

Für das Kyrie und das Gloria mit seinem Tropus hatte er nur eineinhalb Seiten zur Verfügung. Während das Kyrie noch eine geschulte, gotische Schrift zeigt, weist der Abbildung 5. Vollendung einer fehlenden Rubrik im Graduale magnum, A-Gu 17, fol. 334 Schluss des Gloriatropus einen eiligen und modern anmutenden Schriftzug auf, wobei die Federhaltung und die abgerundete Minuskel deutlich das normale Schriftbild des Schreibers verraten.

128 129 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Abbildung 10. A-Gu 17, fol. 257

Abbildung 8. A-Gu 17, fol. 234v

Letzteres, ein späteres Offizium, konnte ich bisher insgesamt in drei Seckauer Hand- schriften finden: im o.g. Graduale A-Gu 17, im Liber ordinarius A-Gu 1566 und auf- Umso interessanter zeigte sich in den Nachträgen des Liber ordinarius A-Gu 756 auf grund der darin vorkommenden Zitierung im sogenannten Graduale minus, das nicht fol. 116 (Abbildung 9) jene Schrift, die vermutlich im Graduale A-Gu 17 auf dem Blatt mehr erhalten ist (Abbildung 11). 234 zu sehen ist. Bei A-Gu 756 handelt es sich um den früheren Liber ordinarius von 1345, den Huebner mit großer Wahrscheinlichkeit nicht nur kannte, sondern ihn sogar dem liturgischen Stand vor 1600 anpasste.

Abbildung 11. Messformular mit Folioangabe des Graduale minus in A-Gu 1566, fol. 209 Abbildung 9. A-Gu 756, fol. 116

Mit Hilfe zahlreicher Folioangaben, die in A-Gu 1566 vorkommen, war es möglich, Im vorgelegten Nachtrageteil des Graduale, der sich unmittelbar vor dem Sequentiar einen partiellen, virtuellen Index der liturgischen Gesänge des nicht mehr erhaltenen findet, wurden auch die Ölweihe des Gründonnerstags sowie das Offizium De Passione Graduale minus anzufertigen. Bei dem Vergleich der zwei Gradualien anhand der an- Domini mit dem Introitus Humiliavit se Iesus eingetragen (Abbildung 10). gefertigten Indices erwies sich der Grad der Übereinstimmung trotz Unvollständigkeit – wie erwartet – als hoch.

130 131 CANTUS PLANUS 2014 MIKLÓS — DER SECKAUER LIBER ORDINARIUS VON CA. 1595 (A-GU 1566)

Wie bereits zuvor erwähnt, überliefert A-Gu 1566 wichtige Informationen über die im Schlussfolgerungen Seckauer Dom zelebrierten Messen. Wir wissen, dass bei der Frühmesse das Graduale Mit welcher Absicht der Liber ordinarius A-Gu 1566 verfasst wurde, kann man nur minus und bei der Hauptmesse das Graduale magnum verwendet wurden. Das zitier- in enger Verbindung mit der Person Georgius Huebner – exzellenter Kenner des Be- te kleine Graduale am Freitag vor Palmsonntag sowie das angegebene Messformular standes der Stiftsbibliothek und Sakristei sowie der Seckauer Liturgie seiner Zeit – be- ohne Alleluia und Sequenz würden für eine Frühmesse sprechen. Jedoch lässt die Be- trachtet werden. Zahlreiche Eintragungen in mehreren Seckauer Quellen vermitteln merkung «post tertiam in sacrario» (fol. 209) vermuten, dass die Messe Humiliavit se den Eindruck, dass er ein eifriger, ambitionierter Gelehrter war. In den süddeutschen Iesus kurz vor 1600 als prior und nicht als Hauptmesse (wie das vollständige Formular Diözesen, die den Römischen Ritus übernommen haben, war der Prozess der Über- im Graduale A-Gu 17 zeigt) gefeiert wurde. nahme langwierig und dauerte nicht selten ein Jahrzehnt (Daschner 1995, 634-635). Huebner, als ein wahrer Archivarius, dokumentierte im Seckauer Liber ordinarius Die auf das Messrepertoire fokussierten Untersuchungen zeigten folgende Ergebnisse: A-Gu 1566 den letzten Stand der mittelalterlichen lokalen Chorherrenliturgie unmit- telbar vor der Übernahme des Römischen Ritus in Folge des Trienter Konzils. Die Der Liber ordinarius A-Gu 1566 stellt eine Differenzierung der Messfeier dar. Wie liturgische Ordnung der musikalischen Gattungen im Liber ordinarius A-Gu 1566 schon oben erwähnt gibt es unterschiedliche Formulare für Frühmesse, Sacrum ist eindeutig mittelalterlich. Die Beweggründe, weshalb Huebner auf die alte Form und Hauptmesse; die Differenzierung betrifft sowohl die Messformulare als auch zurückgriff, bleiben unklar. Vielleicht versuchte er, das Rad der Zeit zurückzudrehen. die dabei verwendeten Bücher. Ihm ist es zu verdanken, dass die spätmittelalterliche Liturgie und Musik gründlich dokumentiert, das liturgische Erbe gerettet und die Wesensmerkmale seiner Zeit für Obwohl in mehreren Fällen dasselbe Proprium für unterschiedliche Messen am die Forschung greifbar gemacht wurden. Festtag vorgeschrieben ist, gibt es eine feine Differenzierung bei den vorgeschrie- benen Ordinariumsgesängen. Während die Ordinariumsgesänge immer den- selben Text haben, sind bestimmte Ordinariumsmelodien mit einem gewissen Festrang oder Grad an Feierlichkeit verknüpft. Die Musik bestimmt hier die Li- turgie (Praßl 2013, p. 255). So weisen an einem Festtag die Ordinariumsmelodien der verschiedenen Messen in A-Gu 1566 große Unterschiede auf, wie folgendes Beispiel zeigt: das Fest der Leidenswerkzeuge Christi ist im Graduale A-Gu 17 auf jener Folioseite mit einer eigenen Kyriemelodie De Lancea (Melnicki 1995, 97) notiert, die auch in A-Gu 1566 erwähnt wird. Weil diese Melodie mit einem etwas jüngeren Fest verknüpft ist, kommt sie erst im Nachtragsteil des großen Graduale A-Gu 17 auf fol. 255, direkt im Formular des Hochfestes De Lancea vor. Das Formular des Sacrum verwendet in A-Gu 1566 hingegen die Kyriemelodie Paschale. Umso feierlicher wirkte das Summum bei den Seckauer Kanonikern, da die Melodie De Lancea nur einmal im Kirchenjahr gesungen wurde. Diese Kyriemelodie ist in mehreren deutschen Handschriften überwiegend aus dem fünfzehnten Jahrhundert ent­halten,16 darunter auch zwei Südtiroler Graduale, die in Brixen aufbewahrt sind.17

16. D-Mbs Clm 9508 von 1452, fol. 283v; CH-SG 546 von 1507, fol. 38v; CH-SG 392 aus dem sechzehnten Jahrhundert, 75 u. a., vergleiche Melnicki (1955, p. 104).

17. I-BREd 2 und I-BREmd sine signatura, vergleiche Melnicki (1955, p. 104).

132 133 CANTUS PLANUS 2014

Bibliography

Amon Karl (1960), Die Steiermark vor der Glaubensspaltung. Kirchliche Zustände 1490-1502, Graz: Styria (Geschichte der Diözese Seckau 3).

Behrendt Inga (2009), Der Seckauer Liber Ordinarius von 1345 (A-Gu 756). Edition und Kommentar, Phil. Dissertation, Kunstuniversität Graz.

Daschner Dominik (1995), Die gedruckten Messbücher Süddeutschlands bis zur Übernahme des Missale Romanum Pius v (1570), Frankfurt am Main: Lang (Regensburger Studien zur Theologie 47). III Jungmann Andreas (1962), Missarum sollemnia. Eine genetische Erklärung der Römischen Messe, 2 vols., Wien-Freiburg-Basel. Editing medieval monophony:

Landwehr-Melnicki Margaretha (1955), Das einstimmige Kyrie des lateinischen Mittelalters, Regensburg: between the old and new philologies Gustav Bosse (Forschungsbeiträge zur Musikwissenschaft 1).

Miklós Réka (2016), Der Seckauer Liber Ordinarius A-Gu 1566. Edition und Kommentar, Phil. Dissertation, Kunstuniversität Graz (online abrufbar: gams.unigraz.at/o:cantus.seckau3).

Praßl Franz Karl (1987), Psallat ecclesia mater. Studium zu Repertoire und Verwendung von Sequenzen in der Liturgie österreichischer Augustiner-Chorherren vom 12. bis zum 16. Jahrhundert, Theol. Dissertation, Graz.

– (2013), ‘Pavlov’s Dog. Listening and Recognition in Gregorian Chant’, De musica disserenda 9/1-2, pp. 253-269.

134 135 Andreas Haug

Towards a semiotically informed transcription practice

1. The current standard method of transcribing melodies from historical staff notations, that is, as found in medieval manuscripts, into modern ones is a procedure so common and so commonly approved that the question of how the melodic content encoded in medieval notations of monophonic music can be most adequately rendered in scholarly editions does not appear to be a matter of dispute, much less a matter of reflection. When we apply what I have just labeled the «current standard method of transcription», we transform the plurality of different historical notations from different periods and regions into a single, more or less unified and stable modern one. We are all familiar with the invariant features of that standard notation: the employment of the five-line staff and octave treble clef; the rendering of pitches with stemless black note-heads; the rendering of neumatic groupings through spacing and/ or with slurs; and the rendering of liquescence, oriscus, quilisma, and strophici with particular symbols. Some will also indicate the isolated virga and the virga-element of compound neumatic figures by adding stems, some will use slurs for distinguishing pitch symbols graphically connected by a continuous pen stroke within neumatic figures from those graphically disconnected, or simulate suchgraphic continuity by leaving no gaps between the respective note-heads in the transcribed passages. Some of us prefer to use another form of modern notation as the medium for our transcriptions: dating back to the nineteenth century, it consists of square notation on four lines with c- and f-clefs. As different as it may sometimes seem on the surface, it is in fact, as we will see, quite similar to the standard notation in any theoretically relevant historical or semiotic aspect.

2. Obviously, when applying the current method of transcription, we detach the melodic content from the cultural contexts and conceptual conditions of past modes of music-writing so as to record it within a coherently constructed, thus unambiguously comprehensible, yet, at the same time, alien and alienating scientific medium: a notation that belongs only to the present, only to us, and which seems to stand outside history, outside the very history of which the transcribed music has been a part. We turn from the astounding variety of different notations, notions that participate in the singularity of the manuscripts containing them, manuscripts in the essential sense of the word, to a notation that shares the typographic stability and rigor of printed books, books being exact replicas of each other, something medieval manuscripts never were. How can such inattentiveness to the distinctive feature of the individual manuscript

137 CANTUS PLANUS 2014 HAUG — TOWARDS A SEMIOTICALLY INFORMED TRANSCRIPTION PRACTICE that its notation irrefutably is, be consistent with the attention contemporary philology with the aim justifying it theoretically, rather than suggesting another method of pays to the materiality and singularity of the medieval handwritten book? And how can transcription. As an adherent of a mainstream practice – a practice challenged only such radical suppression of notational diversity be consistent with the general esteem by a decreasing minority within the scholarly community – one could easily get the of difference in postmodern scholarship? Furthermore, is the standard notation we wrong impression that one is exempt from providing reasons for this practice, or use for our transcriptions as extra-historical and historically neutral as it pretends to from responding to arguments that have been raised against it. The core argument of be? Is, after all, the current method of transcription a procedure as innocent, trivial, those rejecting the current method of transcription – as recently expressed by Manfred and theoretically unproblematic as it appears to be? At least a minority of scholars Hermann Schmid (2012) – is the following: in contrast to the graphic signs of script, – mainly German speaking – have always held the view that it is not. Shouldn’t we, which refer to phonetic signs that in turn refer to meaning, the graphic signs of musical hence, have good reasons for the way we choose to transcribe medieval melodies? I notation directly refer to musical meaning. As a result of this allegedly unmediated think we should, and I think we do. relationship between visual sign and audible sound in musical notation, notation had to follow the changes that music had been subjected to. For that reason, according to 3. Both the current standard transcription practice and the most crucial arguments the Munich school, the specific kind of notation originally belonging to a specific kind against it seem to be of German origin. The wide and lasting acceptance of what became of music cannot be replaced by another notation without a substantial loss of historical the current practice seems to be a result of the influence of German musicology and information. Therefore, instead of detaching music from its ‘own’ notation by replacing the impact German emigration had on North American musicology. But at the same the original shapes of notational signs with modern ones, the only valid method of time there is also a scholarly tradition in German musicology that categorically rejects transcription is to faithfully reproduce the historical shape in handwriting or print – to transcription into modern notation. The founder and foremost protagonist of that provide a Nachschrift instead of an Umschrift. This argument – combining historical tradition was Rudolf von Ficker (1885-1954), a student of Guido Adler and himself a and semiotic concerns – deserves attention, although one could also ask exactly what professor – first in Vienna, from 1927 to 1931, and then in Munich, from 1931 to 1948 information actually gets lost in standard transcription. One could further ask whether (Christensen, Drexel, and Fink 2012). His ideas concerning what he considered the the gap between the music and its ‘original’ notation might not be as deep or even essential and indivisible relationship between music and notation and his belief that deeper than the gap between the original notation and modern transcription. Isn’t that no notation of the past can be substituted by any other without a grave loss of historical the case when music from originally unwritten traditions is only later recorded in information had little lasting influence on later scholarship. This lack of influence notation? When the extant sources of a corpus are much later than the corpus itself? Or resulted from political circumstances and the editorial policies of state-operated when the same corpus of melodies has been transmitted in a large variety of diverse German research institutions during the Third Reich. One of von Ficker’s fundamental notations from different periods and cultural or intellectual contexts? In the material contributions concerning the issues of music, notation, and transcription seems to our community mainly deals with, these are highly frequent cases. have been effectively suppressed and only recently re-discovered (Lindmayr-Brandl 2012). After the war, his suggestions were still being broadly ignored by mainstream 4. But if we set all these reservations aside, what above all makes the assumption of a German musicologists, and transcription into notation with modern features was even direct relationship between sign and sound problematic is the fact that notational signs pompously propagated as «German Standard Technique» (Albrecht 1954, col. 1122: in medieval manuscripts actually do behave like signs, that is, they function under «die deutsche Standardtechnik der Gegenwart»). That his ideas have not been entirely conditions to be perceived and to be described in semiotic terms. Taking distinctions extinguished from the collective memory of the scientific community is due to his once stressed by Charles Sanders Peirce (1933, p. 423) or as proposed by Nelson student Thrasybulos Georgiades (1907-1977), later successor to the Munich Lehrstuhl Goodman in the famous chapter entitled The Theory of Notation in his Languages of (Schick and Erhard 2011). Georgiades vigorously propagated his teacher’s convictions, Art (1976, pp. 127-173) as a point of departure, these conditions can – for our modest although with little resonance beyond the circle of what one may call the Munich purpose and without any claim for theoretical consistency – be described like this: school (Georgiades 1977, pp. 73-80: «Sprache, Musik, schriftliche Musikdarstellung», the notational signs we see in front of our eyes in a medieval manuscript do not relate and pp. 107-120: «Musik und Schrift»). On closer inspection, all of these historical directly to the melodic content they denote. The actual signs, that is, the single visual circumstances might turn out to be a remarkable series of occurrences within the inscriptions or marks produced by the stroke of pen are ‘instances’ or ‘tokens’ of ‘types’, history of our discipline, and future research may still need to determine whether or or, as Goodman prefers to say, of ‘characters’. Characters are defined as «classes of not the omnipresent octave treble clef truly deserves its sinister reputation as ‘Nazi inscriptions or marks» (Goodman 1976, p. 131). Characters must be disjoint classes of clef’. Yet, instead of delving deeper into history, politics, and polemics, let me return inscriptions. This demand is an essential feature of notations: «No mark may belong to to the present and reassess the current practice of transcription in theoretical terms, more than one character», and no two inscriptions must «collapse into one character»

138 139 CANTUS PLANUS 2014 HAUG — TOWARDS A SEMIOTICALLY INFORMED TRANSCRIPTION PRACTICE

(p. 133). This demand is imposed by the primary duty the characters fulfill: in our it fails to satisfy the demand imposed by the duty of a transcription: namely, it fails to case, the duty to identify melodic content from one performance to another. Goodman clarify the relationship between types and tokens. By failing to perceive the individual (pp. 134-135) is aware that marks as tokens related to types the mere depiction of marks fails to properly ‘read’ them as ‘signs’. Moreover, if the simulation is performed by hand, the replicas are not we have in the world not a realm of inscriptions neatly sorted into clearly separate classes but, exact, and the transcription produces new tokens in addition to the historical ones. rather, a bewildering miscellany of marks differing from each other in all ways and degrees. Mimetic copying retains – or even increases – ambiguity rather than removing it. Or the To impose a partitioning into disjoint sets seems a willful even though needful violence. And 1 […] there will almost inevitably be many marks for which it will be difficult or even virtually transcription replaces the multitude of tokens with a stable form of the respective sign. impossible to decide whether or nor they belong to a given character. This stable form then represents a notational type, a ‘disjoint class of inscriptions’. Such an edition satisfies the demand imposed by the duty of a transcription no less and no «A bewildering miscellany of marks differing from each other in all ways and degrees»: more than any edition employing modern note shapes does.2 Does that not sound familiar to anyone who has ever transcribed melodies from medieval manuscript notations? And the decision, whether or not marks belong to a given character experienced as an act of «willful even though needful violence»: Isn’t transcribing just such an act? Transcription implies the decision, whether or not a notational sign belongs to a given character, in other words, whether or not a token is a token of a given type. This decision is based on our understanding of how the notation in question works. The signs we use in the transcription are equivalents to types, not to tokens. The transcriptional equivalents denote the melodic content encoded in the notational types.

5. From these preconditions follow requirements to be met by any semiotically informed transcription. The first and basic requirement concerns the relationship between token and types: the transcription has to clarify which tokens are interpreted as belonging to what types. The second requirement concerns the relationship between the notational types and the transcription equivalents: the transcription has to determine this relationship. It has to bridge the gap between the notational types of the manuscript and the transcriptional equivalents. Thus, the transcription has to be based on distinctive transcriptional equivalents referring to unambiguously defined notational types, rather than to single instances or tokens of those types (that is, to individual marks or inscriptions). Furthermore, the transcription has to include a sufficiently detailed description of the original notation. This description must provide a record of the notational types (as defined by the editor) related to the transcriptional equivalents rendering these types (as chosen by the editor), and a documentation of the tokens interpreted (by the editor) as belonging to a given type. This documentation has to be considered as an indispensible element of any transcription that can claim 1. This is the method employed by Thomas Kelly and Mathew Peattie in their forthcoming edition of the repertory of Beneventan Chant in the series Mmmae. to be both semiotically and historically informed. By including it, the editor takes responsibility for the editorial decisions made in the process of transcription, belittling 2. The forthcoming edition of medieval Latin monophonic music currently being prepared at Würzburg University employs the ‘current standard method’ of transcription: five-line staff and octave treble clef; rendering of pitches with stem-less black neither the historical, nor the semiotic problematics of that complex process. note-heads; rendering of neumatic groupings through spacing; slurs for distinguishing pitch symbols graphically connected by a continuous pen stroke within neumatic figures from those graphically disconnected; rendering of liquescence, oriscus, quilisma, and strophici with distinctive symbols. The transcriptions are accompanied by detailed descriptions of relevant 6. From the foregoing deliberations, it has become clear that the editorial method features of the historical notations dealt with, including the repertory of original notational signs and the transcriptional equivalents employed. We hope that the reasons for our editorial choices are clear from the reflections presented in the of simulating the original shapes is based on a semiotic misconception. Either the present paper. See Corpus monodicum. Die einstimmige Musik des lateinischen Mittelalters, im Auftrag der Akademie der transcription attempts to provide replicas of individual notational marks, at which point Wissenschaften und der Literatur, Mainz, am Institut für Musikforschung der Universität Würzburg, Projektleitung Andreas Haug, www.adwmainz.de/projekte/corpus-mnodicum-die-einstimmige-musik-des-lateinischen-mittelalters-gattungen.

140 141 CANTUS PLANUS 2014

Konstantin Voigt Bibliography

Albrecht Hans (1954), Editionstechnik in Mgg, vol. 3, coll. 1110-1146. Reconstructing acts of writing Christensen Lukas - Drexel Kurt - Fink Monika eds. (2012), Rudolf von Ficker (1886-1954). Tagungsband Editorial consequences of writing scenarios assumed zum Symposium anlässlich seines 125. Geburtstages und des 85-jährigen Bestehens des Innsbrucker Instituts für Musikwissenschaft, Innsbruck: Innsbruck University Press. for the versus Annus novus in Paris 1139

Georgiades Thrasybulos (1977), Kleine Schriften, Tutzing: Schneider.

2 Goodman Nelson (1976 ), Languages of Art. An Approach to a Theory of Symbols, Indianapolis-Cambridge: Dealing with written records always involves the reader’s imagination about how those Hackett. records were made: was the scribe of a particular record working from an exemplar or Lindmayr-Brandl Andrea (2012), Rudolf von Fickers Einleitungstext zum sechsten Auswahlband der Trienter from memory? Was he faithfully copying or «joyfully» varying his model (Cerquiglini Codices in den DTÖ. Ein wieder aufgefundener Entwurf aus dem Nachlass in Christensen, Drexel and Fink 2012, pp. 103-140. 1999, p. 78)? Was he sketching possibilities or providing representative copies? In drastic cases, the different scenarios envisaged drive different assumptions about the L[orenz] K[uno] (1996), Type and Token in Jürgen Mittelstrass ed., Enzyklopädie Philosophie und Wissenschaftstheorie, 4 vols., Stuttgart-Weimar: Metzler, vol. 4, pp. 359-360. compositional principles and the modes of transmission of pieces and repertories (Treitler 1974, pp. 333-374; 1982, pp. 11-60), and more generally speaking, such Peirce Charles Sanders (1933), Collected Papers, 8 vols., Cambridge, Mass: Harvard University Press, vol. 4. assumptions (often tacitly) determine editorial decisions. This contribution, therefore, Schick Hartmut - Erhard Alexander eds. (2012), Thrasybulos G. Georgiades (1907-1977). Rhythmus - Sprache is an appeal for awareness of and reflection on the premises we make about the acts of - Musik: Bericht über das musikwissenschaftliche Symposium zum 100. Geburtstag in München, 1.-2. November 2007, Tutzing: Schneider. writing and their consequences for editing medieval monophonic music.

Schmid Manfred Hermann (2012), Notationskunde. Schrift und Komposition 900-1900, Kassel: Bärenreiter. The shift of editorial philosophies caused by new philology – from author to scribe, from archetype to variant, and from critical edition to source edition – has given new weight to questions about the making and status of each particular written record, because the role of the manuscript source has changed. Once perceived as being just one witness to an ‘original’ text that had to be reconstructed, each manuscript’s version is now appreciated as a text in its own right.1 But even if all versions are considered of equal value, not all notational records can automatically be considered to be correct. This leads us to the main question that stands behind all editorial decisions: how is difference evaluated and represented? How can an editor react to differences among manuscript versions, among strophes of a strophic song, between expected musical conventions and the written record, or between an exemplar and its adaptation? What kind of role do our assumptions about the act of writing play for editorial decisions about difference and sameness, correctness and faultiness?

The strophic versus Annus novus from the manuscript F-Pn 11392 provides an apt example: the written record of this ‘declination-hymn’ with polyphonic refrain displays striking variation among the melodies of the five strophes, while at the same time standing as a model of clarity in writing. The notation is a typical example of the

1. See Cerquiglini (1999, throughout, especially pp. 77-78). Moreover, it is a well known fact that medievalist disciplines did practice a kind of new philology avant la lettre in many respects. A classic example is found in Stäblein (1956), a source edition of hymnaries for the Mmmae series. A change in the editorial philosophy for late medieval German texts was introduced prior to new philology by Kurt Ruh’s Research group for late medieval prose in the 1970s. See Steer (1985, pp. 37-52).

2. Fols. 36v-37; see http://gallica.bnf.fr/ark:/12148/btv1b6000946s/f80.image.r = latin%201139.langDE.

142 143 CANTUS PLANUS 2014 VOIGT — RECONSTRUCTING ACTS OF WRITING means that the first main scribe of the oldest section of F-Pn 1139 had for providing strophe to unison in the following strophes. In strophe 3, the transposition of the first a diastematically unambiguous display of a small range melody in the authentic half of verse 3 (from a-f-g-e to g-e-f-d) melodically disturbs the sequence of thirds and d-mode. The drypoint-line in the lower middle of the notational field above each text seconds that characterizes the line in its other instantiations. This variant seems to line is defined as f by the frequently occurring rhombic mi-signs, and confirmed by the be related to the unison-connection of verse 2 and 3 under the condition of the new consistent position of the final d at the end of each strophe.3 Moreover, a custos at the cadential figure e-f-g at the end of verse 2. The other, large transposition in strophe end of every line secures the diastematically unambiguous continuation of the melody 3 reverses the situation: the second half of verse 1, which is g-e-f-d in most cases, across the line break. This notational clarity allows an unambiguous transcription of appears not lower, but one step higher as a-f-g-e. Once again, this shift destabilizes the the pitches notated by the scribe: tonal syntax of the strophe. The open ending of verse 1 on the pitch e creates a relation between this and the following second verse that is different from the characteristic Example 1. Annus novus, F-Pn 1139, fol. 36v-37, synoptic transcription of the strophes leap from the final d as the last pitch of verse 1 to the fifth a as the opening pitch of verse 2. A variant of syntactical relevance also occurs at the end of verse 3: while the first two strophes descend to the pitch e that belongs to the contrasting secondary third- chain in d-mode, the last three strophes emphasize the primary third-chain with the leap from a to f. Thus, rarely are the variants in this record of Annus novus a matter of different surface realization of the same underlying tonal disposition; instead, almost every variant signifies a syntactical change in the melodic structure.

Three different writing scenarios could serve as explanations for this striking constellation of variants that transform the tonal outline of the generally strophic song. Scenario A was implicitly suggested by Wulf Arlt (1986, p. 39). He assumed that the written record in F-Pn1139 reflects the process of re-formulating the melody while notating it: «einen Prozeß des Formulierens bei der Niederschrift». According to Arlt, this process can be understood as «Zurechtrücken aufgrund eigener Vorstellungen» (ibidem) – as adjustments the scribe made according to his own ideas about the melody. Several records of Benedicamus songs in the Norman-Sicilian manuscripts The pitches and the arrangement of the strophes in this transcription are identical E-Bn 289 and 19421 are witnesses to such modal and syntactical modifications of to what Wulf Arlt provided in his important article on structure and performance melodies during the process of copying.4 This in turn emphasizes the plausibility of of the Nova Cantica (Arlt 1986, pp. 37-44). The first strophe is notated with text. For Arlt’s scenario – a scenario implying the assumption that the scribe used an exemplar the following strophes, the melodic variants are indicated with letter notation. This that was diastematically clear to him. In the case of Annus novus, the scribe would synoptic arrangement shows not only the high degree of variation in general, but also have faithfully copied the first strophe and then have modified it gradually until he the reoccurrence and the singularity of variants. In twelfth-century manuscripts it is arrived at his own version in strophe 4, which he then repeated as strophe 5. This not surprising to find variants among different strophes of the same song (Hiley 1981, scenario raises questions as to the status of the written record: while strophe 1 seems pp. 808-823). But the variants in Annus novus go beyond the expected. In contrast to to represent the melody of the exemplar, and strophes 4 and 5 the scribe’s finished the more typical cases of strophic variation in twelfth-century song, the variants in reformulation, the strophes in between appear to be most transitional, like a kind Annus novus are not only internal to the line, but they also affect line endings. Thus, of compositional sketch. They show the path leading from an initial shape to a final they change the tonal structure of the overall strophe. The scribe wrote three different formulation, without necessarily being self-contained or equivalent to the model or its cadential gestures for the second verse: e-f-g, e-g-a and f-g-a. These varying cadential adaptation. pitches alter the relationship of verses 2 and 3, from an ascending tone in the first

3. Leo Treitler (1967, pp. 70-77) was the first to realize the systematic connection between the modes and the pitch- 4. Wulf Arlt (1970, p. 188) has explained this for Gaude Ierusalem in E-Mn 289, which is subject to an almost complete definition of the drypoint line in the repertory of versus in F-Pn 1139. During the transcription work for a new edition modal restructuring. A less drastic but highly representative example for what I call dynamic variation (as opposed to of the twelfth-century sources of the repertory of ‘New Song’ most of Treitler’s modal assignments proved to be correct, equivalent variation) is found in the record of Lux omni festa in E-Mn 19421, where the process of continuous reworking although they were only based on mode and range. In many cases they could be verified (or falsified) by considering the of the melodic formulations can be traced up to the level of characteristic signs for characteristic variants. See Voigt (2016, mi-signification that Marie Noel Colette (1990, p. 301) discovered in the records of the versus. pp. 301-304).

144 145 CANTUS PLANUS 2014 VOIGT — RECONSTRUCTING ACTS OF WRITING

Three kinds of editions could be based on scenario A (precise exemplar, first copied, ‘final version’ in strophes 4 and 5. If creating a modally smoother, more d-f-a-based then modified): melodic version was the scribe’s intention in the modifications of his model (Arlt 1986, p. 39), strophe 3 comes as a surprise. It reverts to the first version of the second line, Edition no. 1 would only present the result of the scribe’s compositional process: it it brings in yet another secondary third-chain ending on e at the cadence of verse 1 would use the melody of strophes 4 and 5 for all strophes and record the process of and a beginning on g (instead of a) for verse 3. The fact that two of the variants are arriving at that melody in the apparatus and commentary. This edition would assume transpositions of whole segments by a tone and relate to modifications that had been that tonal instability could not have been the scribe’s musical intention, but rather the made in strophe 2 might suggest that they could well be the result of graphic problems. path to a valid version of the tune. The transposition in verse 1, syllable 5-8 could have resulted from maintaining the interval structure (third, second, third) from strophe 1, but applying it to the beginning Edition no. 2 is what Arlt and Dominique Vellard chose for the ‘sonic edition’ on on a, which comes from the corresponding variant in strophe 2. The reappearance of the Nova Cantica CD (1990). They set all strophes with the melody of strophe 1, the cadence e-f-g at the end of verse 2 in the third strophe might not even have been the hypothetical exemplar, which Arlt (1986, p. 39) considered to be «komplexer noticed by the scribe, as he continued verse 3 on g instead of a, as if he was writing the und spannungsreicher» – more complex and with greater tension than the variants variant he created in strophe 2. Regaining his concentration at the end of verse 3, the of the other strophes. A comparison of the first and last strophes soon reveals what scribe decided to solve a problem not previously addressed, and thus he removed the Arlt meant by this. In the first strophe, lines 1 and 4 clearly articulate the primary e at the cadence of the line in favor of f, a solution he kept until the end. In summary, third chain. They start on the final or its fifth, descend into the final at the verse for some variants, one is tempted to find justifications on the level of the scribe’s cadence, use either the fifth or the peak tone c in the third-chain d-f-a-c as the turning sonic intentions; others that are musically less plausible to us – in particular the point, and set up a modal framework in which the internal lines 2 and 3 function as transposition-variants of strophe 3 – might be accounted for as purely graphic writing- contrasts. Both of the contrasting lines 2 and 3 have open endings on the secondary mistakes. Yet, such classifications of the variantsmust remain highly hypothetical. chain of thirds c-e-g-b, which provides tension towards the cadence on the final in Therefore, they would take place on the level of apparatus and commentary, not in the line 4. Moreover, the last two notes of line 2 and the first six notes of line 3 resume established text. In the case of edition no. 3, the established text is solely a rendering of the melody of line 1 almost exactly, despite overlapping the verse boundaries. The what the scribe actually wrote. melody of strophes 4 and 5 does not exhibit these elaborate features. The pitches of the secondary third chain are eliminated from the crucial points at the beginning or Two of these editorial solutions, editions nos. 1 and 3, do also support a different end of lines and the correspondences across the line structure are reduced by the writing scenario. This scenario B assumes an exemplar that was less diastematic than modified cadence of verse 2. Once again, the situation seems to resemble that of some the notation used by the scribe of F-Pn 1139. The variants could then be explained pieces in the Norman-Sicilian sources, in which the more specific and uncommon as a process of reconstructing the tune from a notation that the scribe himself found melodic versions of the first strophes were sacrificed for modally more conventional ambiguous. This approach gains its plausibility from the fact that some of the variants ones in the following strophes. Thus, edition no. 2 preserves the version which is the are shifts of complete melodic segments up and down by seconds. Such shifts point to most interesting and therefore supposed to be closer to the ‘original’. Yet it is a strong coherence in the gestures, but also to uncertainty about the actual pitch level. The fact statement about the estimation of the scribe’s work not to edit his final version, but the that F-Pn 1139 was written around 1100 supports this scenario, especially since the melody of a hypothetical exemplar. scribe in question himself produced diastematically uncertain records in some cases, suggesting that he was educated in a tradition in which adiastematic and diastematic Edition no. 3 would represent all variants found in all the strophes, based on the music writing coexisted and interacted.5 Edition no. 1, as justified by scenario B, would assumption that all steps in the transformation process are equivalent versions. be understood as the successful result of a reconstruction process from an unclear However, it is hard to believe that all of the variants actually do equally represent exemplar. Edition no. 2 is not possible within scenario B, because the melody intended the scribe’s sonic-intentions, considering the differences in character that occur by the exemplar would remain as ambiguous for us as it was for the scribe. In scenario between the variants in strophes 2 and 3. Strophe 2 can be described in terms of modal clarification: in verse 1, the continuation on syllable 5 after the grammatically altered 5. Est hodie rex glorie, the piece directly following Annus novus in F-Pn 1139, shows work of the same scribe in a very words Annus novus (Anni novi, Anno novo etc.) is shifted from the secondary third- different writing context. The mi-significations are ambiguous, and the vertical positions of pitches are extremely unstable. A transcription of the piece according to the principles of the transcription of Annus novus would result in three modally chain’s g to the primary chain’s a, and the open ending of verse 2 on g is removed varying strophes with the same melodic gestures for each line. Hypothetically, the difference between the two records could in favor of the modally more stable a. This latter variant also made its way into the be traced to writing with and without an exemplar. For example, one could assume that the scribe gave a more ‘neumatic’ written expression to a piece he knew by heart, and a more diagrammatic one to the piece he copied.

146 147 CANTUS PLANUS 2014 VOIGT — RECONSTRUCTING ACTS OF WRITING

B, Strophe 1 would then not be ‘the melody’ of the exemplar, but a first attempt our Conclusion scribe made to transfer the graphic information of an ‘imprecise’ record of a melody Editing is never the same as transcribing. Establishing the text of a manuscript version into his ‘precise’ notational diagram. Thus, edition no. 3 could be justified by the always demands decisions about which differences are considered equivalent variants assumption that even in a first attempt the scribe ouldw have reconstructed the melody and which are assumed to be of another status. These decisions can differ radically, just within the framework of his conventions of musical correctness. Strophe 1 would then as the scenarios that support them contradict one another, but still provide plausible not represent ‘the version’ of the exemplar, but our scribe’s understanding of what he explanations of the written record. Facing the variety of parameters involved, a modern found in his exemplar, before he started altering it according to his own ideas in the edition cannot always give the one and only correct solution, but it can document and successive strophes. Commentary should discuss which variants might be due to the reflect the criteria that led to the decisions taken. As I hope to have shown, the notions scribe’s sonic intention and which could be signs of his struggle with the graphical we have about how a written record came into existence are decisive for the kinds of aspects of the notation. The edition itself should represent the process as it stands, decisions we make: for example, if we assume a scribe to be sloppy, would we trust the because in notations of ‘dynamic variations’ scribes are naturally concerned with both results of his writing process, or would we look for the traces of a (better) exemplar in aspects, seeking the best solution in sonic and in graphic representation of the tune them? On the other hand, if we do not assume the existence of a written exemplar, the intended. category of mistakes resulting from the scribe’s uncertainties in reading is an invalid one. If we see notation not as copying, but as a «vicarious performance» in the sense of Other scenarios do not involve a written exemplar at all. Scenario C would treat the Leo Treitler, would we treat the ‘modal’ instability of the record of Annus novus as the setting in F-Pn 1139 as the written record of a highly variable performance. As the scribe’s sound intention, or would we have to reconsider our assumptions about the written record is unambiguous in terms of notational preciseness, the editorial solution preciseness of the notation? would once again be edition no. 3 (all variants); naturally considering the possibility that mistakes may have been made not only in vocal performances, but also in their As it is often difficult to determine under which conditions a musical record was made, «vicarious» (Treitler 1982, p. 49) written counterparts. Scenario D would evaluate the it is crucial to document how we assume that it could have been made as a framework of notated record as a document of the compositional process, and could therefore result the editorial decisions. For a paleographically and/or musically complex record, such in edition no. 1 or edition no. 3, depending upon the editor’s interest either in results or as the one discussed, it is even desirable to give more than one solution in combination in processes. Table 1 summarizes the four scenarios discussed. These scenarios are – with a precise rendering of the assumptions that led to the different editions. The users to make this very clear – intended as models for evaluating a notated record in relation of editions – scholars and musicians alike – will profit more from gaining insight into to its possible production conditions, but not as exclusive explanations of the writing the complexity of transferring historical manuscript notations into modern ‘scores’ processes – which we will in many cases never be able to tell from the manuscripts and the variety of plausible options that this process sometimes reveals, than from a they produced. text that covers up the many plausible solutions by the doubtful certainty of just one version an editor decided for.

Table 1. Writing scenarios and editions

Scenario A ScenarioScenario A B Scenario B ScenarioScenario C C ScenarioScenario D D Precise exemplar, Imprecise exemplar, Written record as Written record Precise exemplar, Imprecise exemplar, Written record as Written record copied, then modified gradually variable performance as act of formulating copied, gradually ‘reconstructed’variable withoutperformance written as actthe ofmelody formulating without then modified ‘reconstructed’ without writtenexemplar the writtenmelody exemplar without exemplar written exemplar Edition 1 Scribe’s last version

Edition 2 Version of the hypothetical exemplar

Edition 3 Edition 1 Edition 2 Edition 3 All variants Scribe’s last version Version of the All variants represented equally hypothetical exemplar represented equally

148 149 CANTUS PLANUS 2014

Elaine Stratton Hild Bibliography

Arlt Wulf (1970), Ein Festoffizium des Mittelalters aus Beauvais in seiner liturgischen und musikalischen Working realities of the New Philology: Bedeutung, Köln: Volk. considering the potential of technology in editing – (1986), ‘Nova cantica. Grundsätzliches und Spezielles zur Interpretation musikalischer Texte des Mittelalters’, Basler Jahrbuch für historische Musikpraxis 10, pp. 13-62. medieval texts

Cerquiglini Bernard (1999), In Praise of the variant. A critical history of philology, Baltimore: The Johns Hopkins University Press. Bernard Cerquiglini’s 1989 publication Éloge de la variante, translated in 1999 as Colette Marie Noel (1990), La notation du demi-ton dans le manuscrit Paris, B.N. Lat.1139 et dans quelques In Praise of the Variant, quickly became a central feature in an ongoing scholarly manuscrits du Sud de la France in Claudio Leonardi - Enrico Menesto eds., La tradizione dei tropi liturgici. Atti dei convegni sui tropi liturgici Parigi (15-19 ottobre 1985) - Perugia (2-5 settembre 1987), discussion of how to edit medieval writing. The book was welcomed by some as Spoleto: Centro Italiano di Studi sull’Alto Medioevo, pp. 297-312. a harbinger of new editorial practices and the banner of a ‘New Philology’; others

Hiley David (1981), The liturgical music of Norman Sicily. A study centred on manuscripts 288, 289, 19421 responded more skeptically, questioning both how new and how beneficial this new and Vitrina 20-4 of the Biblioteca Nacional, Madrid, Ph. D. dissertation, University of London King’s philology actually was.1 Welcomed and criticized, the work proved to be successful: College. the editorial priorities and ideals articulated by Cerquiglini are accepted and shared Nova cantica. Latin songs of the high middle ages (1990), Dominique Vellard and Emmanuel Bonnardot, by many in the current generation of scholarly editors. Deutsche Harmonia Mundi, CD. RD77196 (Schola Cantorum Basiliensis Documenta RD77196). Some of the central points in Cerquiglini’s influential and provocative publication are Stäblein Bruno (1956), Hymnen. Die mittelalterlichen Hymnenmelodien des Abendlandes, Kassel: Bärenreiter (Mmmae 1). well-known: «Medieval writing does not produce variants; it is variance» (1999, pp. 77-78). The author argues that the differences among versions of the same ‘text’ in Steer Georg (1985), Textgeschichtliche Edition in Kurt Ruh ed., Überlieferungsgeschichtliche Prosaforschung. Beiträge der Würzburger Forschergruppe zur Methode und Auswertung, Tübingen: Narr (Texte und separate manuscripts are a direct result of a writing process – foreign to our modern Textgeschichte 19), pp. 37-52. mentality – that relished «an endless rewriting» and «a joyful appropriation» of existing material (ibidem, p. 78). The differences among multiple manuscript versions are Treitler Leo (1967), The Aquitanian repertories of sacred monody in the eleventh and twelfth centuries, Ann Arbor, MI: Umi Research Press. not to be explained away, suppressed, or buried in a critical apparatus. Rather, since «the variance of a medieval work is its primary characteristic, […] editions must give – (1974), ‘Homer and Gregory. The transmission of epic poetry and plainchant’, The musical quarterly 60, pp. 333-374. [variance] a priority» (ibidem). Each individual manuscript version is to be presented as valid and complete. – (1982), ‘Observations on the transmission of some Aquitanian tropes’, Forum musicologicum 3, pp. 11-60. Less well-known is Cerquiglini’s proposal of how to do this. How can one produce Voigt Konstantin (2016), Chancen und Grenzen des überlieferungsgeschichtlichen­ Paradigmas für eine Edition des ‘Neuen Liedes’ in Dorothea Klein ed. Überlieferungs­geschichte trans­disziplinär. Wiesbaden: editions that celebrate and display the «joyful excess» of variability in medieval Reichert (Wissensliteratur im Mittelalter 52), pp. 283-306. writings (ibidem, p. 21)? How can an editor avoid anachronistic presentations of ‘texts’ and ‘authors’? According to Cerquiglini (p. 79) an ideal edition of medieval writing can hardly be achieved within the format of a book. He states instead: «The computer provides the obvious solution». The assertion bears repeating: Cerquiglini claims that when editing medieval writing, it is not the book, but rather the computer, that provides an appropriate medium, an obvious solution. Yet in 1989, at the time Cerquiglini wrote, the computer was only a tantalizing promise, a potential to be dreamed about. For certain medieval texts – particularly those transmitted in a large number of manuscripts – the type of edition promoted and described by Cerquiglini

1. Ziolkowski (2005) provides the most comprehensive overview of the scholarly context and reaction to Cerquiglini’s work (as well as the complex relationship between the ‘new’ editorial philosophy and editorial practice). In this review, Ziolkowski references the central contributions of the 1990 Speculum 64/1 issues (including Nichols). See also Wenzel (1990); Stackmann (1998); Bennewitz and Weichselbaumer (2005).

150 151 CANTUS PLANUS 2014 STRATTON HILD — WORKING REALITIES OF THE NEW PHILOLOGY could not be achieved at the time of his writing. Cerquiglini, in essence, offered editors because «computer inscription [like medieval writing] is variance» (ibidem, p. 81). The a Herculean task – prioritize no individual manuscript version of a text above another computer screen, in its flexible, ephemeral, transitory, «mobile» presentation, is the – and stated that they would soon have a tool that would allow them to complete it most appropriate platform for presenting medieval writing (ibidem, pp. 81-82). easily – the computer. In Praise of the Variant describes a utopia where the computer has rescued editors from their most difficult problems. With the benefit of twenty-five intervening years, we are in a better position to evaluate these claims. Today, some of the computer’s early promises seem less promising. The My contribution examines this lesser-known aspect of Cerquiglini’s influential work: storage capabilities of a computer are certainly superior to those of a book, but they its understanding of the benefits technology might provide editors of medieval writing. themselves can become a disadvantage. An abundance of un-interpreted information I will also consider whether the technological tools Cerquiglini promoted are indeed (what Cerquiglini imagined as a purely positive offering of technology) has itself capable of the editorial feats he imagined. The twenty-five years that have passed since become a problem. The limiting factor to how much information can be conveyed is the publication of In Praise of the Variant allow for a more nuanced evaluation of the no longer the book, it is now the human: realistically, how much information can an author’s claims. In 2014, we are better able to determine how the promise of editing editor meaningfully organize and provide? How much information can a reader-user with technology compares with the reality of experience. We are able to acknowledge meaningfully process? A flood of information can become a desert of knowledge. A the positive aspects of using technology foreseen by Cerquiglini; we are also able to second disadvantage of the digital realm concerns the role of the editor. Technology engage cautiously with aspects of the digital realm that have proved to be problematic. provides a potential substitute – and in my opinion, an unsatisfactory substitute – for This presentation is also an attempt to re-evaluate the usefulness of Cerquiglini’s the editor. I pose the following questions as an impetus for further conversation: does publication for the field of chant scholarship. For while the book attracted enormous the benefit of an editor end when all manuscript images and indices of manuscript attention among those working with medieval texts, especially texts in vernacular contents are available on-line? Is accessibility to information and images enough? I languages, the book has had less impact among musicologists. One notable exception suggest that even while developing and using technological tools we maintain our occurred with the paper given by James Borders at the 1990 gathering of Cantus editorial role – our ability to interpret medieval writing and analyze transmission planus; his paper explored how the central tenets of the New Philology might address patterns. Cerquiglini himself imagined that the role of the editor would be facilitated, particularities in the transmissions of tropes in the Italian repertory (Borders 1992). not eliminated, by the use of technology. The most serious challenge associated with the use of technology is providing for the long-term storage and accessibility of What exactly did Cerquiglini hope to achieve by using computers – digital technology research data. According to one – necessarily coarse – estimate, of the digital projects – to edit medieval writing? His work outlines five advantages. First, a computer funded by the Dfg (Deutsche Forschungsgemeinschaft, the national funding body for screen might provide a dialogic user-interface. The reader could become a «user» and German research grants), only about fifty percent are still accessible to outside users.2 acquire control over how the work of the editor is viewed (Cerquiglini 1999, p. 79). This suggests that approximately half of the completed projects are now inaccessible, Second, Cerquiglini considers the ‘multidimensional’ screen to be superior to a two- except to those researchers who created them. Such an estimate should sober those dimensional page, given the computer screen’s ability to retrieve, eliminate, maximize, of us who attempt to create a digital edition to replace a book. The costs and labor and minimize displays of information through the use of windows (ibidem). As a involved in keeping user interfaces and digitally stored research data accessible are third advantage over the book format, the computer offers a vastly superior storage ongoing and should not be underestimated. capability. A computer can spare the editor the «painful choice» of selecting what information to include and what to exclude (ibidem, p. 80). Using a computer, an editor I conclude with some considerations of the advantages that technology might provide can accommodate every type of information imaginable and «satisf[y his] need to be an editor of medieval chant. These goals are humble, compared with those offered by exhaustive» (ibidem). Fourth, Cerquiglini thought that a computer format might allow Cerquiglini, but they seem achievable and offer tangible benefits for scholars. First, the knowledge of the editor to be more readily accessible for the reader-user. Far from a dialogic user interface that allows a user to determine how the editor’s work is eliminating the need for an editor, technology, according to Cerquiglini, could provide presented offers a meaningful research tool. Users of a digital edition could be able to a way of presenting editorial readings and editorial decisions in a clearer way. The role select edited texts – even those occurring in separate, printed volumes – and arrange of the editor is preserved and her work is visible in the type of digital edition imagined them on the screen in a preferred order, according to the user’s individualized criteria. by Cerquiglini. The fifth advantage of technology described by Cerquiglini is also his most daring assertion: Digital editions might change our conception of medieval 2. Torsten Schrade (Akademie der Wissenschaften und der Literatur Mainz, Digitale Akademie), personal communication writings as ‘texts.’ The new medium – the digital medium – has the potential to move via email, 7 July 2014. Publications concerning the long-term digital storage of research data, such as Neuroth, Strathmann, Oßwald, Scheffel and Klump (2012), and Neuroth, Oßwald, Scheffel, Strathmann and Huth (n.d.) offer forward-looking us away from inappropriate understandings of fixed texts and authorial intentions, guidelines and goals rather than statistical analyses of completed projects.

152 153 CANTUS PLANUS 2014

This would allow for an efficient comparison of the variants among concordant versions of a chant. Also, the increased storage capabilities of the digital realm might be profitably harnessed for editions of certain genres of plainchant – such as the sequence – that have transmission patterns so extensive that they exceed the capacities of a printed volume. Using increased storage capabilities brings increased labor, but one possibility for reducing the additional editorial work could come in the form of collaboration. Contribution mechanisms, such as those used in the Cantus database and websites,3 might allow scholars – the users of the digital editions – to provide their own transcriptions to the digital platform. The most interested users of a scholarly IV edition might also participate in the editing. This is an advantage of the digital realm Chant and computing not imagined by Cerquiglini, and perhaps a fitting one with which to end: the use of technology allows the international community of scholars to collaborate more easily across geographic distances, creating resources that are both mutually produced and mutually beneficial.

Bibliography

Bennewitz Ingrid - Weichselbaumer Ruth (2005), Lob der Variante(n)? New Philology und die Praxis der mediävistischen Editionen in Christa Jansohn - Bodo Plachta eds., Varianten - Variants - Variantes, Tübingen: Niemeyer, pp. 61-77.

Borders James (1992), Tropes and the New Philology in László Dobszay - Ágnes Papp - Ferenc Sebó eds., Cantus Planus: Papers Read at the Fourth Meeting; Pécs, Hungary; 3-8 September 1990, Budapest: Hungarian Academy of Sciences, Institute for Musicology, pp. 393-406.

Cerquiglini Bernard (1999), In Praise of the Variant: A Critical History of Philology, transl. Betsy Wing, Baltimore-London: The John Hopkins University Press; or. ed. Éloge de la variante: Histoire critique de la philologie, 1989, Paris: Éditions du Seuil.

Neuroth Heike - Strathmann Stefan - Oßwald Achim - Scheffel Regine - Klump Jens eds. (2012), Lang­ zeit­archivierung von Forschungsdaten:­ Eine Bestands­aufnahme, Boizenburg: Hülsbusch.

Neuroth Heike - Oßwald Achim - Scheffel Regine - Strathmann Stefan - Huth Karsten (n.d.), nestor Handbuch: Eine kleine Enzyklopädie der digitalen Langzeitarchivierung und Langzeitverfügbarkeit digitaler Ressourcen für Deutschland, Boizenburg: Hülsbusch.

Nichols Stephen G. (1990), ‘Introduction: Philology in a Manuscript Culture’, Speculum 65/1, pp. 1-10.

Stackmann Karl (1998), Neue Philologie? in Jens Haustein ed., Philologie und Lexikographie: Kleine Schriften, Göttingen: Vandenhoeck & Ruprecht, pp. 20-41.

Wenzel Siegfried (1990), ‘Reflections on (New) Philology’, Speculum 65/1, pp. 11-18.

Ziolkowski Jan M. (2005), ‘«Metaphilology». The Powers of Philology. Dynamics of Textual Scholarship by Hans Ulrich Gumbrecht. Error and the Academic Self. The Scholarly Imagination, Medieval to Modern by Seth Lerer’, The Journal of English and Germanic Philology 104/2, pp. 239-272.

3. University of Waterloo, Cantus: A Database for Latin Ecclesiastical Chant, [website]. (Accessed on 24 February 2015); available from http://cantusdatabase.org/ and http://cantus.uwaterloo.ca/

154 155 Debra Lacoste - Jan Kolá ek

Cantus for office and mass: building an online network of chant databases

The freedom with which information is now exchanged on the Internet has promoted new methods of research involving interactive databanks and online image libraries. Cantus: A Database for Latin Ecclesiastical Chant is one of the oldest computer- based research projects devoted to medieval chant, having been conceived in the mid- 1980s.1 Its multiple rebirths over the years have encouraged scholars in diverse fields to both use and contribute to the database, and it continues to be an indispensable research tool for musicologists, liturgists, art historians, and scholars in other fields. A brief history of the project reveals three main stages: 1) It was created at the Catholic University of America in the late 1980s using mainframe computers with Cobol applications, Gopher protocol for file transfers, and floppy diskettes sent in the post. Each manuscript index was a separate file in a fixed ‘.txt’ format of eighty characters per line, readable on any monitor without the need for special graphics features. 2) In 1997, the project was transferred to Western University in Ontario, where the individual files were entered into a database. Index records were produced offline and then converted through Asp to be available on the Internet in Html. 3) In 2011, Cantus was transferred to a Drupal server at the University of Waterloo (Ontario, Canada) in order to manage the large quantity of records (nearly 400,000) now stored in the database. New analytical tools were created, and online data-entry and record management have since become available. This transformation from a static database to a highly-interactive and flexible web-based format has brought the programming behind the user interface up-to-date and on par with many of the research websites currently on the Internet.2

Cantus is well known as a resource of digital inventories or ‘tables of contents’ of the chants contained in manuscript and early printed sources of the medieval liturgical office. The interactive nature of the website allows for a variety of searches and the collection of data, and is an ideal instrument for: a) identifying chants either in medieval service books or that may have been used as cantus firmi in polyphonic motets; b) placing liturgical texts that may have been embedded in sermons; c) providing evidence of the provenance of a source; d) analysing selected chants within regional or widespread traditions. Shortly after announcing five years of renewed funding from the

1. Access Cantus at http://cantus.uwaterloo.ca (accessed January 10, 2016).

2. The Cantus Database has been hosted at the University of Waterloo since March 2011 as part of the Margot research cluster of medieval digital humanities projects devoted to the Middle Ages (Moyen Age et Renaissance: Groupe de recherche - Ordinateurs et Textes). Access Margot at: https://uwaterloo.ca/margot/ (accessed January 10, 2016).

157 CANTUS PLANUS 2014 LACOSTE - KOLÁ EK — CANTUS FOR OFFICE AND MASS: BUILDING AN ONLINE NETWORK

Canadian government beginning in April 2013,3 Cantus began a new venture: linking under development, one of the aims of Simssa is to provide a central repository for together several different online databases devoted to Latin chant. This project, the digital copies of musical scores with complementary information about their contents. Cantus Index,4 provides a growing central repository of chants that have been found in Simssa’s focus so far has been on early musical sources because they provide a realistic manuscripts for both the mass and liturgical offices. Thus, the Cantus Database has but rewarding challenge for the Optical Music Recognition (Omr) software being built effectively been split into two websites: the ‘database’, which stores the information to process the digital images. Simssa has begun scanning medieval chant manuscripts collected over the past two decades from manuscripts and which offers analytical and applying Omr to transfer the musical contents into a searchable database. As tools for research, and the ‘index’, which is the main textual reference and houses the Optical Character Recognition (Ocr) has allowed people to search for particular words master listing of standardized chant texts. within Google Books, Omr enables access to specific musical information within the original images, with searchable layers of content information. Simssa’s initial ventures Through the adoption of common feast names and our ‘Cantus ID numbers’, which are have demonstrated that despite the hundreds of different types of medieval signs and unique numbers for each chant text, the Cantus Index website enables compatibility the unique characteristics of scribes across medieval Europe, the musical, textual, and among various existing chant databases; we are sharing chant data. Along with the liturgical content on manuscript pages can be isolated and identified. Hours of research manuscript data in Cantus, our partners include: Portuguese Early Music Database,5 time spent collecting data could potentially be saved using this Omr technology, and Slovak Early Music Database,6 Gradualia (Hungarian Chant Database),7 Fontes Cantus musicologists could move towards analysis of the information much more quickly. Bohemiae,8 and Cantus Planus in Polonia.9 With regard to ease of use, a one-click search With a potentially very large amount of digitized chant data extracted with reduced for a chant in Cantus Index returns matches in any and all connected databases, with time and effort, the scope of computer applications for analysis is considerable. The live links directly to those records. Chants listed in Cantus Index contain only four data two main sources that Simssa has used, to date, as proofs-of-concept are the sixteenth- fields: full text, genre, feast (not required), and Cantus ID number. Registered users century Salzinnes antiphoner from the area of Namur, Belgium, now held in Halifax, contribute by adding new chants, which immediately receive automatically generated Canada (CDN-Hsmu M2149.L4), and the well-known early antiphoners copied by the Cantus IDs for subsequent references to manuscript databases.10 Saint Gall monk Hartker (CH-SGs 390/391). Both of these sources are fully indexed in Cantus, and the images for both manuscripts are freely available on the Internet in Not only is Cantus an excellent resource for chant researchers and singers, but it high resolution. also provides the kind of detailed information about important medieval musical manuscripts needed to ensure the success of other projects. While Jan Kolá ek has On their website, Simssa uses Cantus data to provide content information along the approached other projects and invited them to collaborate in our Cantus Index ‘master right-hand side of a screen which is otherwise filled with a digitally-processed image chant’ listing, we, in turn, have been approached by several other online projects that of each manuscript folio. Similar to Cantus’s searching capabilities, Simssa also are eager to collaborate with Cantus. We have entered into an exciting partnership with allows the user to search for particular text or pitch sequences, thus mining Cantus McGill University for a project entitled Single Interface for Music Score Searching and data. Thriving on the wealth of online digital image libraries, where high-quality Analysis (Simssa).11 Simssa is under the direction of Ichiro Fujinaga and is funded by photographs of thousands of pages of medieval books are freely available, Simssa will the same Canadian-governmental agency that funds the Cantus Database. Currently not only complement the digital tools currently accessed in medieval chant research, but will bring their strongest aspects together in a unified online research environment. Cantus’s role in this development has been to provide the manuscript information, 3. Social Sciences and Humanities Research Council of Canada ‘Insight’ Grant (2013-18): ‘Cantus’ for office and mass: New Online Tools for Textual-Melodic Input, Search, and Analysis of Mass Chants. and our project was selected as a partner based on the integrity of our data. The lab

4. Access Cantus Index at: http://cantusindex.org (accessed January 10, 2016). – that is, the dozen or so graduate and undergraduate students – responsible for the

5. Access Portuguese Early Music Database at: http://pemdatabase.eu/ (accessed January 10, 2016). development of Simssa projects is the Distributed Digital Music Archives and Libraries

6. Access Slovak Early Music Database at: http://cantus.sk/ (accessed January 1, 2016). Lab (Ddmal) in the Music Technology area of McGill University. Although Ddmal is

7. Access Gradualia (Hungarian Chant Database) at: http://gradualia.eu/ (accessed January 10, 2016). also working on questions concerning audio digitization and recording techniques,

8. Access Fontes Cantus Bohemiae at: http://cantusbohemiae.cz/ (accessed January 10, 2016). this is the place where a scanned and fully searchable Liber usualis (text and melody) 12 9. Access Cantus Planus in Polonia at: http://cantus.edu.pl/ (accessed January 10, 2016). can be found.

10. At the time of this presentation in August 2014, Cantus Index contained thousands of chants for the office: 11,533 antiphons, 5,282 responsories with 8,664 responsory verses, 927 versicles, and many newly-entered records for mass chants, including 275 introits and 273 communion antiphons. 12. For more information, see: http://ddmal.music.mcgill.ca/research/omr/search_the_liber_usualis (accessed January 11. Access Simssa at: http://simssa.ca/ (accessed January 10, 2016). 10, 2016).

158 159 CANTUS PLANUS 2014 LACOSTE - KOLÁ EK — CANTUS FOR OFFICE AND MASS: BUILDING AN ONLINE NETWORK

Dot Porter, co-chair of the Mellon-funded project Medieval Electronic Scholarly While our new Cantus Index houses thousands of standardized chant texts, the Alliance (Mesa), a federation of digital medieval projects,13 has also approached us. original Cantus Database continues to hold an important place within the research Mesa essentially acts as a cross-search of collections and projects, enabling scholars community with its approximately 400,000 records representing the contents of over and students to simultaneously search metadata and full-text for a variety of materials, 135 manuscripts.14 including transcriptions, edited texts, manuscript descriptions, and museum artifacts. They currently federate metadata from twenty-two projects and collections, including Following is a brief summary of some of the developments in the Cantus project since the British Museum, the Bibliothèque nationale, the Gothic Ivories Project at the the last progress report to the Cantus Planus Study Group: Courtauld Institute of Art, the Digital Image Archive of Medieval Music (Diamm), 1. Cantus moved to the University of Waterloo, a new website was constructed, and and Teams Middle English Texts. They are interested in extending Mesa further into another update is due to be released this summer; music and the Cantus Database made their short-list. 2. The new Cantus Index website was implemented both to store ‘master’ versions of chant texts and to be a central hub through which projects can inter-connect on the web; This past year we were approached by Agostino Paravicini Bagliani, Chair of the Cost 3. Hundreds of mass texts have been entered into Cantus Index by Elsa De Luca to Action ‘Medioevo europeo’ of the organization European Cooperation in Science enable the Cantus Database to accept indices of graduals; and Technology. In April 2014, our web-developer Jan Kolá ek attended a workshop 4. Thousands of records have been entered by users worldwide through the online in Florence where medieval databases containing textual material that could be Cantus Input Tool; developed towards a Medieval Digital Latin Library were identified. The discussion 5. Calvin Bower’s Sequence Database has begun to be incorporated into the site and was begun concerning the technicalities related to the potential interoperability of more than 500 new sources have been added to the manuscript listing; these medieval, Latin archives. The partners for this proposed Medieval Digital Latin 6. Full texts have been entered for all of Paris, Bibliothèque nationale de France Library are the International Society for the Study of Medieval Latin Culture - Società - Département des Manuscrits, lat. 12044 (F-Pnm lat. 12044) in preparation for Internazionale per lo Studio del Medioevo Latino (Sismel) and its digital library, collaboration with Simssa; the Edizione nazionale dei testi mediolatini in Italy (Entmi), and the Archivio della 7. In addition to our team of Debra Lacoste, Jan Kolá ek, and Kate Helsen, one latinità italiana del Medioevo (Aiml). The Sismel and Fondazione Ezio Franceschini undergraduate and two graduate students are working on Cantus at the University of (Fef) Digital Humanities Lab are currently coordinating and developing the first Waterloo; prototype, with the aim being a free cross-consultation of the participating online 8. Two graduate students at Dalhousie University in Halifax (Canada) are updating databases. Such a tool would aid in the development of medieval studies by giving a records in the Cantus Database through their work for Simssa; unique or at least a main reference for quotations and queries about words, names of 9. Full Volpiano melodies are being entered for Paris, Bibliothèque nationale de France persons and places, and concepts across time and countries. - Département des Manuscrits, lat. 12044 (F-Pn lat. 12044) by the Dalhousie students (overseen by Simssa); Jason Stoessel from the University of New England in New South Wales, Australia, 10. Full texts and full melodies have been entered for the two famous Hildegard of is leading a consortium of Australian and international researchers in a proposal Bingen sources, the Dendermonde manuscript (B-DEa 9) and the Riesencodex to repurpose medieval chant data and make it freely available, in particular, the (D-WIl 2). The full contents are awaiting proofreading by Cantus staff, and will then Medieval Music Database of John Stinson that has been dormant for many years. be released into public view. They are calling this new research tool eMirabilis, which stands for ‘Early music 11. Presentations have included annual sessions at the International Medieval Congress Manuscripts, Incunabula and Resources in Australian, British and International in Kalamazoo, MedRen, the Canadian Society of Medievalists, the Florence meeting for Libraries and Internet Sites’. They plan to connect with other online databases, such the Medieval Digital Library project, and for Simssa in Montréal. as Diamm and Cantus, to not only provide exposure for these other websites but also to integrate data and enable easier access to the wealth of digitized information With international partners both contributing to and accessing these websites, that is currently available online, including manuscript images, bibliographies, and Cantus Index and its parent database are taking the next step towards global full-text discographies.

14. At the time of publication of this report, the database held the contents of over 170 manuscripts; this increased number is owing to the contributions of individual scholars worldwide through the online data-entry tool. While these new indices await proofreading by Cantus staff, their ‘draft-form’ contents are incorporated into some of the analytical programmes 13. Access the Medieval Electronic Scholarly Alliance at: http://mesa-medieval.org (accessed January 10, 2016). where they can already be of value for researchers.

160 161 CANTUS PLANUS 2014 searching of chant manuscripts, as well as analysis of individual sources and liturgical Kate Helsen - Inga Behrendt - Jennifer Bain traditions. The Internet, which allows the crossing of time and space in an instant, has profoundly changed how we interact with medieval chant manuscripts. Digital A new way to see neumes: technology has not only allowed us to connect textual data records to images of the the optical neume recognition project original manuscript sources in online archives, but has also enabled uncreased in context and in action perspective beyond the individual liturgical centres where these manuscripts were used. Broad traditions or local customs become evident, and in only a few clicks, we can approach an understanding of the chant that was sung in medieval Western Europe. When we explore new territory, the answers we get are determined by the questions Until recently, connections among chant research projects were rather scant owing we ask. The questions we ask are determined by what we understand in the first place. to incompatibilities with technologies or concepts. With the help of forward-looking And what we understand is influenced by the way in which we encounter that new database managers, the future of chant research appears to be collaborative and territory – what is omitted, what is emphasised, what is assumed. A decade ago, as a networked online. graduate student of chant, my goal was to understand the work that had been done in the field, learn from it, and fall into line using the methodologies and repertories that had stood the test of time. But even then I could sense that the ‘context’ of older studies was different from my context, meaning that my possibilities for research were different. This was mostly due to computers and their application to various aspects Sitography of chant studies: databases (Cantus1 and others), digital transcriptions using word Cantus Database processors, and networking with other research projects through the Internet. But http://cantusdatabase.org computer-aided techniques were still based on the traditional repertory; databases http://cantus.uwaterloo.ca (accessed January 10, 2016) and transcriptions had simply altered the medium through which they arrived on my Cantus Index desk for analysis. http://cantusindex.org (accessed January 10, 2016)

Cantus Planus In Polonia But what if computer technology could help us change the landscape of chant studies http://cantus.edu.pl/ (accessed January 10, 2016) itself and increase the number of sources available, the degree of detail to which we could study them, and the speed at which our questions and comparisons about that Ddmal. Distributed Digital Music Archives and Libraries http://ddmal.music.mcgill.ca/research/omr/search_the_liber_usualis new landscape could be answered? The two key elements that define this change are (accessed January10, 2016) digital images and optical music recognition (Omr). The number of online, high-quality

Fontes Cantus Bohemiae digital images of manuscripts is increasing every day, and the software that can ‘read’ http://cantusbohemiae.cz/ (accessed January 10, 2016) and then encode this information is keeping pace. The Optical Neume Recognition project, first introduced to this group three years ago, is based on the advances of both Gradualia. Hungarian Chant Database http://gradualia.eu/ (accessed January 10, 2016) image processing and Omr software. Using this project as a case study, we can see that the main influence of computer technology on chant studies has been to transform the Margot. Moyen Age et Renaissance: Groupe de recherche Ordinateurs et Textes. https://uwaterloo.ca/margot/ (accessed January 10, 2016) kind of data available to study, making new methodologies appropriate and useful. New questions produce answers that, in turn, challenge previous assumptions about Mesa. Medieval Electronic Scholarly Alliance http://mesa-medieval.org (accessed January 10, 2016) what it is possible to know about chant, and what uncharted avenues of inquiry are worth pursuing. Pem. Portuguese Early Music Database http://pemdatabase.eu/ (accessed January 10, 2016) In a lecture on the disconnect between post-modern and empirical approaches to Simssa. Single Interface for Music Score Searching and Analysis musicology, David Huron (1999) suggests that the choice of one methodology over http://simssa.ca/ (accessed January 10, 2016) another has less to do with a scholar’s philosophical approach as it does with the

Slovak Early Music Database http://cantus.sk/ (accessed January 1, 2016) 1. http://cantus.uwaterloo.ca/ (accessed January 10, 2016).

162 163 CANTUS PLANUS 2014 HELSEN - BEHRENDT - BAIN — A NEW WAY TO SEE NEUMES: THE OPTICAL NEUME RECOGNITION PROJECT materials available to the scholar. We need not assume that some methods belong Having gained a sense of context, let us now turn to the Optical Neume Recognition to science and others belong to the humanities; our approaches must be tailored to project.3 Originally, the project included only Inga Behrendt and myself, and our the nature of the thing studied. As a historical discipline, the orientation of chant idea started at the Dobogökö Cantus Planus meeting in 2009. A year later, our first scholarship is necessarily a retrospective one: we concentrate on what has been computer scientist, Alan Sexton, of Birmingham University joined us. Since 2012, our left to us and its significance. Our research materials are fixed; we cannot devise an group has grown to include Jennifer Bain (Dalhousie University), Ichiro Fujinaga, experiment to gather new data, in the way other fields might be able to. Pre-computer- and several researchers at his music software lab (McGill University). The project is age (and downright Herculean) accomplishments such as Hesbert’s Cao (Corpus currently funded by the Social Sciences and Humanities Research Council of Canada Antiphonalium Officii) have acted as a kind of basis upon which the occasional new and recently received word that it will be generously funded again by Sshrcc, as part find might be measured, or in view of which certain traditions could be grouped. of a larger endeavour entitled the Single Interface for Music Score Searching and But compared with other ‘brands’ of musicology such as ethnomusicology or music Analysis Project (Simssa)4 for the next 7 years! The immediate goal of the project is psychology, chant scholars have less source material to work with; codices with music to produce a neume editor that is capable of scanning the digital images of Hartker’s notation and early prints, the facsimiles thereof, or even descriptive library catalogues, antiphoner (CH-SGs 390/391) and classifying each neume on the page. To do this, Inga are available in indisputably limited numbers. Even within the chant repertory we Behrendt, Anton Stingl (a freelance musicologist) and I drew up a table containing think of as standard, some traditions might be thought of as peripheral based on their examples of each neume form found in the antiphoner, as drawn by as many hands as origins or historiography. Changing chant studies from a ‘data-poor’ to a ‘data-rich’ we could identify, resulting in a ground truth set of about 500 signs made up of about field means altering something about the existing data itself, since we cannot produce 100 different signs and their respective modifications through liquescents, etc. After more of it simply by financing new experiments. That change is possible through each digital image has been processed by a series of tasks which reduces the visual digitization. noise and enables the neumes to be more easily identified (called binarization), the neume editor works by comparing what neumes it finds on the manuscript page to High-quality digital images of whole manuscripts are appearing in increasing numbers the neume table provided. This process is hosted online, which means that it is not on the Internet, and many of them are available for free. Even more manuscripts necessary to download any software onto one’s own computer – one can conduct this have been digitized but still remain the property of their libraries; access to these can procedure from anywhere with an Internet connection. be bought for a fee and then downloaded by the researcher. The advantage of these digitized images is not only portability, storage, or colour rendering – if this were the Once each neume has been identified (a procedure which, for now, still involves an case, then it would be only a change in the medium, and not in the data itself –, but the error correction process) information about its type, positioning, and relationship to the ability to ‘scale up’ the business of processing them; encoding their content digitally Latin syllables below it is encoded in an Xml schema. These schemas are probably best in an accurate and searchable way. This is analogous to the effect of Google Books on thought of as the computer’s way of describing all the graphic components that make historical literature. Without adding to the corpus itself, digitizing every page of millions up an image. The particular Xml schema used here is called the Musical Encoding of books has allowed new questions to be asked of the literature itself. Searching for Initiative (Mei) because even though it was originally developed to describe modern particular words, detected by optical character recognition (Ocr) software, allows for a musical scores, other programmers and musicologists have expanded its capabilities swift comparison of a vast number of books. But while printed words, and the alphabetic to include other notation types.5 The information about every single neume contained characters that make them up, pose no significant challenge to modern image processing in Hartker’s antiphoner, encoded in Mei, will then be in a format that will allow other software, musical notation is a different story. Optical music recognition (Omr) is chant sources (rendered in Mei) to be compared to it. This means that all the musical indeed comparatively efficient at understanding modern printed musical notation, but information is now not only searchable but may be compared with the musical content processing hand-written notations on historical documents is where the real challenge of other manuscripts, as part of this growing networked environment. For example, I lies. Computer scientists, in collaboration with musicologists, are currently working might ask of one source: «Is this particular cadential gesture always rendered with the on a Networked Environment for Music Analysis (Nema)2 specifically to include early, same choice and order of neumes?» and then I could expand this question to include handwritten sources of music by developing software to identify and process neumes using classifiers, and encode that musical information in a standardized way so that it 3. For updates on the current state of the project, please see our blog: https://opticalneumerecognition.wordpress.com/. can be compared to other digitized images of sources. 4. http://simssa.ca/ (accessed January 10, 2016).

5. One such expansion was lead by Stefan Morent (2008) in his edition of Hildegard of Bingen’s music; see his Digital Musik Edition project at the Universität Tübingen at http://www.dimused.uni-tuebingen.de/hildegard_e.php (accessed 2. https://code.google.com/p/nemadiy/ (accessed January 10, 2016). January 10, 2016).

164 165 CANTUS PLANUS 2014 HELSEN - BEHRENDT - BAIN — A NEW WAY TO SEE NEUMES: THE OPTICAL NEUME RECOGNITION PROJECT other manuscript sources for the same chant repertory. Such queries will involve Bibliography much more data than it is possible to manage manually, and their results will help us to assemble sets of statistical likelihoods about neume choice and context. Huron David (1999), The New Empiricism: Systematic Musicology in a Postmodern Age in The 1999 Ernest Bloch Lectures, Berkley: University of California http://musiccog.ohio-state.edu/Music220/Bloch.lectures/3.Methodology.html Of course, Hartker’s antiphoner is a well-studied source: a core member of the (accessed January 10, 2016). traditional and standard chant repertory with a facsimile edition, Solesmes notation Morent Stefan Johannes - Richert Pfau Marianne eds. (2008), Hildegard von Bingen. Der Klang des studies, and several articles written about it. This is in part why we chose this source Himmels, Wien-Köln-Weimar: Böhlau. to be the proving ground for the concept; our project can produce new kinds of data about a well-known book, and we can compare our data with previous work. Existing theories about neume interpretations can now be tested empirically over large data sets to see if they hold up. We can track and compare the patterns, alterations, and pairing Sitography of neumes to syllables, quickly and accurately. The use and consistency of significative Cantus Database letters in the Saint Gall notation in particular can also be examined. And yet, there http://cantus.uwaterloo.ca (accessed January 10, 2016) may at times be a sense of unease at the prospect of handing over semiological tasks to automated systems. Beyond that, there is the hesitation about applying scientific Digital Musik Edition, Universität Tübingen http://www.dimused.uni-tuebingen.de/hildegard_e.php (accessed January 10, 2016) analysis to something that can also be correctly considered a form of prayer, or at the very least, a precious cultural artefact. The words of David Huron (1999) address this Nema. Networked Environment for Music Analysis https://code.google.com/p/nemadiy/ (accessed January 10, 2016). concern: Optical Neume Recognition Project To the traditional music scholar, it must look for all the world like science is muscling-in on https://opticalneumerecognition.wordpress.com/ musicology. But the rise of empiricism has nothing to do with ‘science’. It arises from within music Simssa. Single Interface for Music Score Searching and Analysis scholarship, and is motivated by the desire to learn as much as possible from the information http://simssa.ca/ (accessed January 10, 2016) available to us – including the additional information that might be assembled with a little effort. […] Changing conditions simply allow us to be better music scholars, to embrace higher standards of evidence, and to be more acutely aware of the moral and aesthetic repercussions of our knowledge claims, including claims that something is unknowable or that some phenomena ought not to be investigated. Our strongest criticisms should be levied at those who insist on speculative discourse when the resources are readily available to test such knowledge claims.

The application of digital technology to these thousand-year-old signs affords them the opportunity to speak for themselves instead of through even the most diligent and experienced interpreters. We can now empirically test what we think we know about them, clearly separating what can be proven from what has been speculated. For example, we can now approach questions about the meaning and occurrence of liquescents, significative letters, cadential gestures expressed as recurring neume patterns, and the relationship between chant text and notation more objectively than before. This new approach may lead us to confirm what has been previously suggested, or to explore new ideas. This methodology respects the source itself as the ultimate authority on what it contains, and investigates those contents on the deepest level. By combining new digital image technologies, optical music recognition (Omr) software, and standardized encoding methods (such as Mei) for the new information gleaned from these processes, the Optical Neume Recognition project is playing its part in creating a better research environment for chant scholars around the world.

166 167 CANTUS PLANUS 2014 MAESSEN — FIRST RESULTS OF A COMPUTATIONAL ANALYSIS OF OLD HISPANIC CHANT

V Alleluias, tropes, sequences, and the mass ordinary

168 169 Gábor Kiss

Late flourishing of the alleluia repertory in central Europe

Concerning the development and extension of the medieval alleluia repertory, some key notions can be found in the introduction to the first volume of Karl-Heinz Schlager’s Alleluia-Melodien (1968, p. vii). Schlager mentions the practice of grouping the alleluia melodies for martyrs and apostles so that they could then be freely chosen for the actual mass formula. From this technique, he reasons, it is only a small step to composing a new melody, or at least, to adding new text to an existing melody. «Mit diesem Schritt», as he concludes, «beginnt jedoch die Geschichte, die das Alleluia im Mittelalter erfahren hat». Already in the so-called basic repertory of the earlier alleluias we can observe an exchange of texts and melodies: the approximately seventy melodies occurring in almost every region are associated with hundred different verses in the sources (Schlager 1965, p. 13). In addition to this standard repertory, we find regional groups of melodies, as well as numerous, sparsely documented alleluias not belonging to any particular region. It is worth noting that the number of regional, local melodies and Einzelmelodien is around five times higher than the number of standard melodies.1 The inner structure of the later repertory appearing in the second volume of the Alleluia-Melodien is different (Schlager, 1987). The 470 new melodies (composed from the twelfth century) were not intended to substitute the core repertory, but rather to supplement it for newer feasts of local saints and additional masses (votive masses). Since this compositional activity reflects different and evolving local liturgical customs, the new chants rarely gained widespread use. According to my calculations, more than half of these melodies, around 280 of the 470, are unique. The majority of chants that can be found in more than one source are confined to a few records from a circumscribed, local environment; of these, only a limited number (eighteen) spread beyond those confines to emerge in a considerable number of manuscripts. Only about ten percent of melodies from the newer repertory are associated with multiple texts, compared with almost a third in the older repertory. However, there is no great difference between the proportion of individual texts and melodies, with the 410 melodies in the old repertory used for 587 different texts and around the 470 melodies in the later repertory used for 600 texts. The reason for this is that adaptations in the old repertory are more evenly distributed and more or less confined to standard melodies. In the late repertory, however, the numerous texts

This research was supported by the Hungarian Scientific Research Fund (K 81763, NK 104426).

1. It must be added, however, that some of the older regional melodies come from the Gallican and Beneventan areas, representing not so much a renewal of the repertory, but rather the preservation of older practices; cf. Schlager (1987, p. 16).

171 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE were applied to a relatively modest number of very popular melodies that were quite sources. Table 3 (p. 183) shows the distribution of melodies according to region, and frequently reused for new texts. Among forty-two melodies having more than one text, table 4 (p. 183) contains melodies that can be documented from two areas. Melodies that only nine ones have around ten or more different verses, whereas the average number appear sporadically in different areas generally belong to religious orders. However, all is two or three (see table 2, p. 183). of the melodies that enjoyed greater popularity and were disseminated geographically in a natural way belong to German, or southern German areas, frequently including We can see that the majority of these popular melodies were originally assigned sources from Bohemia and sometimes Aquileia. The two tables suggest that southern to Marian feasts. This is not surprising, given the surge in new texts and melodies Germany and Bohemia during the late Middle Ages provided lively and stimulating belonging to the cult of Mary among late medieval alleluias. The frequent use of these environments for the creation and reception of new products and for the genre, the melodies for further texts may have something to do with their popularity. They are monodic alleluia itself. all among the most widely used of the melodies, whereas the overwhelming majority of the repertoire is documented singularly, or at best by just a few sources. At the Although the main trends that can be gathered from the earlier works are still valid, same time, dissemination and attractiveness are not by all means mutually inclusive. new evidence suggests that new sources may change our knowledge of concrete There are also widely adopted melodies that were never or rarely subjected to the melodies or the overall picture of the repertory. Examination of a significant number process of creating new adaptations (e.g. Dorsa eorum, for the celebration of apostles of Czech, Moravian, Polish and Hungarian sources (about sixty sources altogether) or evangelists; Ave stillans melle and Felix es sacra virgo for Mary; Qui creavit omnia for that were not included in the complete edition of alleluias has amply justified our Saint Catherine, among many others). Perhaps other reasons lie in the characteristics of expectations. From the sources, many new alleluias and new, earlier and unrecorded the melodies. For example, Letamini in Domino (see table 2) has an easily memorable adaptations came to light (meaning more than 190 new melodies and more than 140 form, consisting essentially of repeated melodic segments (Schlager 1987, p. 677). O adaptations). This is a relatively great number compared to the total number of chants consolatrix and O Maria rubens rosa represent typical melodic strategies of the two in the complete edition of alleluias. Although, many Bohemian sources were used most popular modes of the late repertory, namely the modern, Ionian F mode (unifying in the alleluia edition, new alleluias were found connected to Sigismund (the patron the fifth and sixth mode) and the characteristic liedhafte E-Melodik. saint of Bohemia) and Ludmilla (grandmother of Saint Wenceslas). In the Hungarian sources we have so far uncovered seventy-three new melodies, mainly for Mary The complexity of the alleluia repertory is increased by the fact that, in both the older (thirty), and for the feasts of the local saints Stephen, Ladislaus, Emeric, Adalbert (ten and the later repertory, some verse texts have more than one melody (see table 1, or eleven). In the Polish sources, 107 new melodies have been identified so far: almost p. 183). While thirty percent of the texts in the older layer belong to this category half of them (fifty-two) were composed for Marian feasts, and ten for local saints (five (168 verses), only slightly more than ten percent of the new alleluia texts were for Stanislaus and five for Hedvig). More than sixthy percent of the new melodies are sung to different melodies (seventy-seven). Characteristically, only six of them have unique melodies in both repertories. The difference between the numbers of melodies around ten or more melodies, and they are all assigned to late feasts (Corpus Christi, in the two traditions has something to do with the religious orders. While Hungarian Transfiguratio Domini, Visitatio Bmv and votive masses). It should also be noted that orders generally followed their international cursus, their sources were neglected. In half of the extra-melodies are taken from the standard repertory, so they represent Poland, however, the secular and non-secular repertories are not sharply separated; only new combinations of known texts and melodies. moreover, many local melodies were produced probably in monasteries.

There is also a great difference in the distribution of the melodies by region. As is These statistics show that the vivid, late environment mentioned for the dissemination well-known from Schlager’s introduction to his thematic catalogue, the overwhelming and composition of chants, combination of texts and melodies in a new way was much majority of regional melodies in the early repertory originated in southern France and larger than was previously believed or fully explored. In the following, I will give an Italy (Schlager 1965, pp. 15-16). Starting in the twelfth century, the situation changed: overview of the different layers of new findings through concrete examples. All of my a massive new northern repertory evolved, and very few melodies from French and examples come from the lists mentioned above. The simplest way of extending the Italian areas appear in more than one or numerous sources. Contrary to the standard traditional repertory was to create new combinations of known texts and melodies. repertory, there are very few late melodies that pop up simultaneously in French, This phenomenon is well-known from the earlier repertory, but the practice continued Italian and German sources. We find only two chants that were disseminated in each in the late Middle Ages, too. We have found a significant number of such examples region (O consolatrix pauperum and Salve virga florens Aaaron), and there are only among the Polish and Hungarian sources. Although we may regard them as secondary nine melodies in the whole repertory that occur simultaneously in French and German to the new compositions, they share with the latter the task of fitting together the

172 173 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE text and melody: this sometimes may have been automatic, but it sometimes seems it seems relatively frequent. Again, the following are a few of the many examples: to have required a more creative process. A few examples can be seen in the table the melody of the double-verse Pascha nostrum was almost untouchable; it was set of the adaptations. While the verses Video caelos for Saint Stephen protomartyr and to only two other melodies in very small circles of Italian sources. In the Hungarian Vidimus stellam for Epiphany were generally sung to the second mode melody of Dies practice it was replaced by a new style F-mode melody, which occurs in no less than sanctificatus, in the Hungarian tradition (and only there) they were adapted to the sixth seven sources (sometimes together with the general melody, sometimes alone).7 The mode melody of Domine in virtute.2 Similarly, Veni electa mea was sung to the melody practice even reached such prestigious melodies as that of Dies sanctificatus, which of Surrexit Dominus et occurrens, which was generally not used for other texts.3 The was used throughout the year for different feasts with more than fourthy different alleluia Sancte Paule apostole was sung to different melodies in the standard repertory, verses. This melody was replaced in the Missale Notatum Strigoniense with a modern but the melody of Qui confidunt or Veni Sancte Spiritus were not among them. The D-mode melody spanning a large range, which was originally set to the verse Nativitas former was used in the Bohemian and Hungarian sources, whereas the latter can be gloriose virginis.8 Both examples were mentioned and published in the Hungarian found in Polish manuscripts.4 literature (cf. Szendrei 2005a, p. 308), whereas the following example 1 found in a late gradual from Csíksomlyó in Transylvania. As can bee seen, here the verse Veni Sancte The next step was when either a new text or a new melody was composed for the Spiritus was adapted to a new Ionian F-mode melody.9 alleluia. The motive must have been different in these cases. In the first case, new texts were generally required for new feasts and saints. Sometimes this meant nothing Example 1. Alleluia Veni Sancte Spiritus, RO-MCcsm No. 5252, fol. 21 more than taking a common verse and making it proper by inserting the name of the saint, while in other cases, more specific texts were created. There are also transitional solutions between the two. For example: O rex et apostole Stephane for Saint Stephen of Hungary uses the text O consolatrix as a point of departure, around whose vocabulary and formal characteristics new and specific texts were created.5 While many of the new texts were set to new melodies, it was not a rare practice to adapt new, sometimes modern, versified texts to old, or old-style melodies. This is true for the eastern areas, too, where many new text-melody combinations arose. For example: Subveni mater pia for the Commune Bmv was not taken up in the alleluia editions. Its probably thirteenth- century text was sung to the old melody of Surrexit pastor bonus, and as such, was The most important case, of course, is when new melodies are used with new texts. well-known in Bohemian, Polish and Hungarian manuscripts. Another example (of We can see numerous examples in the examined sources. In both repertories there the many) is the probably Bohemian Inventa una pretiosa (for Saint Margaret or de are great differences in the type and quality of the texts and melodies. Among the Virginibus), which had an own melody, but was adapted in a late Polish manuscript to new texts we find both complex rhymed verses either with a regular or an irregular the melody of Iusti epulentur.6 structure as well as simpler prose texts. While advanced poetical forms can be found in Bohemian and Polish manuscripts, which may have originated in university ambience The opposite also occurs, namely when new melodies were added to the good old or in monasteries, in Hungarian sources they are confined mostly to the adoption of texts. Since those texts generally belonged to traditional feasts and had their own some well-known regional chants. At the same time, an interesting late practice can venerable melodies, the reason for experimenting with a new melody may have been be observed here: when the repertory needed to be extended for later feasts and new of an aesthetic nature. They wanted to refresh the repertory and to celebrate even the votive masses, and no available new texts were at hand (perhaps also in the absence of the necessary skills to produce them), preexisting texts were borrowed from elsewhere, traditional, ‘obligate’ feasts with modern new-style compositions. Schlager (1987, p. viii) referred to this as a rare practice among late Polish and Hungarian sources; however, including from other parts of the liturgy and from other genres. Table 5 (p. 184) lists a few such examples. As a point of departure I use Salve mundi Domina from Schlager’s edition (1987, p. 779), where he himself interprets it as a paraphrase of Salve regina, an 2. Cf. H-Efkö Mss. 3b, fols. 24, 32; SK-BRm EC Lad. 3, fol. 14.

3. Cf. H-Efkö Mss. 3b, fol. 46v. 7. TR-Itks 2429, fol. 102v; SK-Sk Ms. Mus. No. 1, fol. 99v; RO-MCcsm No. 5252, fol. 8; H-Bn Fol. lat. 3815, fol. 58; SK-Brsa 4. Cf. PL-KI s. n., fol. 38v. Sign. 67, fol. 58; H-Bn Fol. lat. 3522, f ol. 56; H-Bu A 114, fol. 34v.

5. E.g. H-Efkö Mss. 3b, fol. 108; H-Bn Fol. lat. 3815, fol. 117v. 8. SK-BRm EL 18, fol. 44.

6. P-TAm s. n. iii. 9. RO-MCcsm No. 5252, fol. 21.

174 175 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE example of connecting different genres through poetic texts. There are similar examples concordance is limited to the alleluia, whereas the verse Egregie has its own melody in the Polish tradition; see, for example, Angelus ad virginem subintrans, which uses spanning a larger range, which left behind the eighth mode reminiscences of Virga the first strophe of a sequence A( h 8, no. 51). Finally, there are a great many examples Iesse. In other cases, whole melodies were taken over for the new texts. In the case of old in the Hungarian repertory, in which the text of antiphons and responsories (generally melodies, the melodies were borrowed more or less faithfully, while modern melodies belonging to the same feast) were taken over and employed as an alleluia (see table 5, were adapted much more freely. It is obvious that for one of the most important chants p. 184). Among new alleluias it is very rare for a melody not to have earlier roots, but for Saint Stephen of Hungary (O rex et apostole Stephane), the popular and frequently the manner and extent to which earlier models were adopted can vary greatly. In the texted melody of O consolatrix pauperum was used. So here it was not just a melody, following example, the alleluia for Saint Stephen is still borrowed from an old-style but a tradition of reusing a melody that was taken over (see example 3a-b). While in alleluia (Aemulor enim), whereas the verse already has a proper melody (example 2). this case the adaptation was faithful to the original, there are other cases where the connection is much looser. See, for example, O quam gloriosum est, which omits certain Example 2. Alleluia Aemulor enim, H-Efkö Mss. 3b, fol. 118v melodic segments from the original melody (see example 3c). In other examples, even Alleluia Sancte rex Stephane, SK-BRm EC Lad. 3, fol. 268v the order of recognizable musical units is changed, and in many cases we can hardly separate the contrafacta, the variants, and the rather free exploitation of the elements of earlier materials from one another. The last melody of the example (see example 3d) shows another type of F mode melody that has much in common with the O consolatrix type, although the progression of the two types is different (Sancte rex Stephane, Praesul Adalberte). Finally there are melodies that seem to use melodic formulas from both typical melodic families (Fundata est domus, Pascha nostrum).

Example 3a. Alleluia O consolatrix, H-Efkö Mss. 3b, fol. 101.

Example 3b. Alleluia O rex et apostole Stephane, H-Efkö Mss. 3b, fol. 108

A similar but more complex example is that of the new alleluia for Saint Sigismundus. The alleluia Egregie dei martyr has an incipit similar to the melody of Dulce lignum. Actually, in at least one instance in the Polish repertory, the same text is truly adapted to the whole melody of Dulce lignum.10 However, the Bohemian Egregie dei martyr echoes not Dulce lignum, but rather the modern and popular melody of Virga Iesse.11 We also know of an example (from Saint Gall), where the melody Virga Iesse was truly used for the verse Egregie.12 At the same time, in the Bohemian source the concrete

10. P-TAm s. n. III; cf. Pikulik (1995, p. 138).

11. P-Wru B 1714, fol. 186v.

12. D-SGs 546, fol. 303; cf. Schlager (1987, p. 829).

176 177 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE

Example 3c. Alleluia O quam gloriosum est, TR-Itks 2429, fol. 92v of how different melodies could easily be produced from the same or very similar stereotyped motives and phraseology. Moreover, the last two melodies (see example 4e-f, p. 180) show that not only poetical forms, but also common musical types may connect different genres, namely alleluias, sequences and ordinary chants.

Example 4a. Alleluia Mediatrix vitae datrix, H-Efkö Mss. I. 3, fol. 253v

Example 3d. Alleluia Sancte rex Stephane, H-Bu A 115, fol. 6v

Example 4b. Alleluia Angelus ad virginem, PL-Kk Ms. 42, fol. 43

It is interesting to see how the popularity of melodies led to their frequent use with new texts, thus increasing their popularity and presence in the liturgy, and how this presence and popularity inspired the evolution of a new style. We can observe the evolution of a common musical language, in which many new melodies could be produced easily from a common collection of typical motives and melodic segments, and by assembling them in different ways and orders. The melodies are obviously similar, and their constituent parts always sound familiar, but they still represent independent chants. At the same time, another scenario is also imaginable. According to Janka Szendrei (2005b, p. Example 4c. Alleluia Ave verbum incarnatum, SK-Sk Ms. Mus. No. 1, fol. 53v 111), the order may be the opposite: it is possible that thinking in stereotyped formulas and free treatment of the form may have preceded the formation of popular melodies, which – according to this concept – can be considered as successful assemblages of the variable material. The same development also took place among E-mode melodies. The difference is that here we cannot specify concrete melodies that became as popular and as frequently used for contrafacts as O consolatrix. At the same time, there are many E-mode melodies that are characterized by the same stylistic coherence (use of the same motives, characteristic turns, stereotyped melodic progressions, etc.) as we observed in the F-mode repertory. The melodies in example 4 are a good demonstration

178 179 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE

Example 4d. Alleluia Post partum virgo, PL-Kj Nr. 1267, fol. 41 late repertory on a regional level. Because melodies and texts were circulating and compositional efforts influenced each other, the characteristics of a local repertory can sometimes be understood best by understanding the other repertories.

We can see this in a few selected examples, where the analysis of new sources brought new insight and invalidated earlier statements and hypotheses about the provenance or dissemination of certain melodies. Some of the problems are associated with the insufficiency or inappropriate interpretation of the sources. The only source from the Hungarian heritage Schlager (1987, p. 889) used was the graduale Wladislai (H-Efkö Mss. I. 3), and he referred to it as Bohemian. The reality, however, is more complex, as the graduale is a blend of several influences, which are chiefly Bohemian, but also Example 4e. Prose Illibata mente sana, PL-Kj Nr. 1267, fol. 100 Hungarian, Moravian and Polish:

1. Margaretha, que decreta, a rhymed alleluia, is documented only from this source (fol. 325v). Although apparently a unique melody, it belongs to the family of E-mode melodies, described above. While no Bohemian records of the melody have been found yet, it was frequently used in Polish sources, representing a Polish component of this complex manuscript (Pikulik 1995, pp. 172-173).

2. Pax pie celse deitatis is also an alleluia singularly documented from this gradual (fol. 256v). However, this comes from a misreading of the source, for the correct incipit is Fax precelse deitatis as is confirmed by a missal from Olmütz (CZ-OLu M III 6, fol. Example 4f. Sanctus, PL-Kk Ms. 42, fol. 22v 307v). This is an important finding for two reasons: on the one hand, it shows that the alleluia was not a unique chant; on the other, it may serve as corroborative evidence for the hypothesis of ordering the manuscript from Olmütz (Körmendy 2013, p. 119).

3. Post partum virgo and Prophete sancti predicaverunt had a common melody in the older repertory. The complete edition registers a new melody for both from the graduale Wladislai alone (fols. 13v, 40v; Schlager 1987, p. 407). However, the melody appears in no less than seven further Hungarian sources, showing that it was a stable element within the Hungarian tradition.13 Since no other records of the melody have been discovered yet from other areas, its Hungarian provenance seems probable. If so, it would not be the only instance where the special gradual drew on the Strigonian use. As is well-known, this stylistic layer of the alleluias was primarily popular in south 4. The celebration of local or regional saints was not always confined to given areas, German and Bohemian sources. It can be seen in the alleluia edition, where among and might also have appeared elsewhere, according to changing fashions. Although an seventy-seven melodies occurring simultaneously in German and Bohemian alleluia for Hedwig, the Polish saint (O felix Hedwigis) appears as a unique chant from manuscripts we find twenty-six F-mode and twenty-six E-mode melodies, so the a German source in the catalogue (Schlager 1987, p. 713), it is by no means surprising two modes are evenly represented. On the other hand, among the seventy-five new that at least its text can be found in many diocesan and monastic Polish sources as Hungarian alleluias we find only thirteen E-mode melodies, whereas the overwhelming well. Moreover, according to Pikulik (1995, p. 199), this commune-like text may have majority of the others (forty-five) use mode F, together with the characteristic style been compiled in Poland at the end of the thirteenth century, and in Polish sources, described above. Furthermore, the E-mode melodies appear primarily in manuscripts that belong to peripheral areas of Hungary, and in which a Bohemian and Polish 13. RO-BRbn Ms. I. F. 67, fol. 88v; SK-Sk Ms. Mus. No. 1, fol. 102; H-Efkö Mss. 3b, fol. 100; RO-AJ Ms.IX.57, fol. 72v; SK-Brsa influence can be detected.This demonstrates well why it is necessary to study this Sign. 67, fol. 122v; RO-MCcsm No. 5252, fol. 6v; H-Bn Fol. lat. 3815, fol. 99.

180 181 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE two other melodies were used for it. A similar case is that of O sydus Polonie for Saint Table 1. Statistical comparison of alleluia repertories Stanislaus, which is presented in Schlager’s edition (1987, p. 739), as a unique chant Melodies Verses Melodies with Verses with from a Viennese source whereas it can also be found, this time with the same melody, several verses several melodies in several Polish manuscripts (Pikulik 1995, p. 227). Moreover, in the Viennese source, Early repertory 410 ~587 122 168 the form of the verse has been changed. Late repertory 470 ~600 142 177 5. The alleluia for Saint Margaret (Inventa una pretiosa) is documented from a significant number of Bohemian manuscripts (seven in the alleluia-edition, which may be extended by many others) and from an Augustinian source from Moosburg Table 2. Melodies used frequently for contrafacts (late repertory) (Schlager 1987, p. 673). Since the latter is the earliest one, Pikulik (1995, p. 50) was inclined to accept the monastery of Moosburg as the place of origin. At the same time, Basic text Feast Number of texts Mode it turned out that the chant was well-known in Hungary too; moreover, its earliest 1 Ave benedicta Maria Bmv 12 F source is the Missale Notatum Strigoniense from the first half of the fourteenth century (SK-BRm EC Lad. 3, fol. 241v). In addition, from the fifteenth century the chant was 2 Conversus Jesus ad Mariam Maria Magdalena 14 F used in Poland as well, where besides the above mentioned melody, another one was 3 Felix es sacra virgo Bmv 27 F also used for the same text. In any case, if the new information does not change the question of provenance, it shows that Inventa una pretiosa was not restricted to a 4 Felix es fructu triplici Petrus martyr 12 F narrow environment, but became a truly regional melody. 5 Letamini in Domino Commune martyrum 13 E

6. Sometimes not only can the Polish sources change the situation as it appears in the 6 O consolatrix Bmv 40 F complete edition, but also the opposite can take place. In his monograph on Marian 7 O Maria rubens rosa Bmv 9 E alleluias in Polish sources, Pikulik (1984, p. 113) states that the available sources Dominicus, Augustinus, indicate that Ave benedicta Maria must have been of Polish origin. But Schlager’s 8 Pie pater Dominice 19 F Benedict, Bernard edition (1987, p. 587), appearing a few years later, makes this unlikely: it documents the 9 Virga Jesse floruit Bmv 23 G/F melody from more than forty sources from Germany and Bohemia starting from the thirteenth century, whereas Pikulik’s earliest document is from the fourteenth century.

Table 3. Distribution of melodies according to regions (late repertory) The aim of presenting these examples is far from that of criticizing great and honorable earlier achievements; on the contrary, the authors were aware of the limits of their Spanish French English Italian German Bohemian Total work, and, accordingly, were prudent in their observations. But these examples (together with many others not mentioned here) may throw light on the nature of Single source 18 65 6 41 126 23 280 this repertory and how it was disseminated and grew. Alleluias, new melodies, new More than one 1 8 – 6 61 16 87 texts or a combination of the two, the many new adaptations and combinations of sources texts and melodies show that the creative approach to this part and layer of the liturgy Well-represented – 1 – 1 9 7 18 was much more intense and generalized than previously assumed. The examples range from complex, subtle and genuine compositions to the simplest compilation of texts and melodies serving practical needs, but the conscious or unconscious reuse of Table 4. Melodies documented from two areas (late repertory) earlier materials was common to every part of the repertory. We may regard examples of it as secondary (Rajeczky 1977, p. 228, called them «Randerscheinungen»), but French–German English–French Italian–German German–Bohemian without analyzing the web of their wide-ranging occurrences, it is impossible to fully Number of melodies 4 4 1 22 understand the essence and history of the genre. It therefore seems appropriate to Number of compile a supplementary regional catalogue to the existing editions. well-represented melodies – – – 10

182 183 CANTUS PLANUS 2014 KISS — LATE FLOURISHING OF THE ALLELUIA REPERTORY IN CENTRAL EUROPE

Table 5. Pre-existing texts used for new alleluia melodies Bibliography

Text Feast Source text Genre Feast Prov Pikulik Jerzy (1984), Śpiewy Alleluia o Najświętszej Maryi Pannie w polskich graduałach przedtrydenckich, Warszawa: Akademija Teologii Katolickiej. Salve mundi domina Bmv Salve regina A Bmv D – (1995), Spiewy Alleluia de Sanctis w polskich rekopisach przedtrydenckich: Studium muzykologiczne, Angelus ad virginem Angelus ad virginem Warszawa: Wydawnictwo Akademii Teologii Katolickiej. Bmv Lied ? P Christi Christi (Ah 1) Angelus ad virginem Angelus ad virginem Rajeczky Benjamin (1977), ‘Kontrafaktur in den Ordinarium-Sätzen der ungarischen Handschriften’, Studia Bmv Sequ Annunt. Bmv P Musicologica 19, pp. 227-234. subintrans subintrans (Ah 8)

Nova bella virginum Undec. mill. Nova bella virginum A Undec. mill. P Schlager Karl-Heinz (1965), Thematischer Katalog der ältesten Alleluia-Melodien, München: Ricke.

Omnipotenti Deo – (1968), Alleluia-Melodien I bis 1100, Kassel-Basel-London-New York: Bärenreiter (Mmmae 7) Bmv de Nive Omnipotenti Deo filii A Bmv de Nive H immensas – (1987), Alleluia-Melodien II ab 1100, Kassel-Basel-London-New York: Bärenreiter (Mmmae 8) Veni Sancte Spiritus De S. Spir. Veni Sancte Spiritus A De S. Spir. H Szendrei Janka (2005a), ‘Egy középkor-végi dallamstílus jelentkezése az alleluia-műfajban’, Zenetudományi Tu vera sapientia Dolgozatok 2004-2005, pp. 107-146. Veni sapientia rex et Votiv Cantio Trin H (Ah 21) – (2005b), A ‘mos patriae’ kialakulása 1341 előtti hangjegyes forrásaink tükrében, Budapest: Balassi Suscipe verbum virgo Bmv Suscipe verbum virgo R Bmv H Kiadó.

Hodie Christus Hodie Christus Nat Am Nat H natus est natus est

Glorificamus te sancta Bmv Glorificamus te sancta A, An Bmv H

Maria intacta virgo Bmv Maria intacta virgo A Bmv H

Tribus miraculis Bmv Tribus miraculis Am Ep H ornatum (Oct.Nat) ornatum Hodie beata virgo Hodie beata virgo Purif. Bmv Am Purif. Bmv H Maria Maria

Regina caeli letare Bmv Regina caeli letare A Bmv H/P

O quam suavis est Corp O quam suavis est Am Corp H

Mons Gargane Mons Gargane Mich. Ang. A Mich. Ang. H noli timere noli timere Benedic Domine Benedic Domine Ded. Eccl. A Ded. Eccl. H domum domum

184 185 Giulia Gabrielli

Tropes in cantus planus sources from South Tyrol

This article aims to introduce the repertoire of tropes in South Tyrolean medieval music manuscripts.1 This topic is closely related to a broader systematic research project into music manuscripts in South Tyrol carried out over the last five years by the University of Bolzano/Bozen and devoted to music manuscripts with cantus planus and cantus fractus from the eleventh to the end of the nineteenth centuries. Research has brought to light the existence of altogether 120 manuscripts, many of them hitherto unknown to historical or musicological research. Among these, medieval manuscripts represent a substantial part, corresponding to around one quarter of the total. The liturgical- musical sources held in Bolzano (seventy-six manuscripts with musical notation) are described in the catalogue edited by the author (Gabrielli 2015).

South Tyrol is a region in the Alps between Central Europe and the Mediterranean area. The region now forms part of Italian territory, but its culture and language were originally Germanic. Until 1964 the region was divided into different dioceses (see the map in Gozzi 2001, p. 469). The main city, Bolzano, belonged to the diocese of Trent (in turn part of the patriarchate of Aquileia), while Bressanone/Brixen was the episcopal seat of a vast diocese belonging to the ecclesiastical province of Salzburg. The westernmost part of the region, instead, fell within the territory of the diocese of Chur/Coira until 1819. The ancient abbey of San Candido/Innichen in Pustertal, which was founded in 769 in what is now South Tyrol, fell under the jurisdiction of Freising. Scattered throughout the region are a number of important churches and monasteries with rich archives and libraries. Medieval music manuscripts are still preserved in Bolzano in the Franciscan Convent (I-BZf, two manuscripts), in the Benedictine Monastery of Muri-Gries (I-BZb, three manuscripts) and in the City Museum (I-BZmc, three manuscripts). A larger number of music manuscripts have survived in Bressanone. Nine medieval music manuscripts are preserved in the Theological Library (I-BREs), with two in the Diocesan Museum (I-BREmd, one in the Diocesan Archive (I-BREd), one in the Vinzentinum private school (I-BRE), and one in the library of the Capuchin Convent (I-BREc). In Neustift/Novacella, a famous Augustinian abbey near Bressanone (I-NV), four manuscripts are preserved. One manuscript can be found in San Candido (I-SCAcc), two in Castle Tyrol, and one in a

1. This article reflects the state-of-the-art in the study on medieval music manuscripts in South Tyrol in 2014. Since then, the research has continued, leading to important results, for example the study by Gionata Brusa on the Liber Ordinarius of Bressanone (second half of the thirteenth century), preserved in the Museum of San Candido/Innichen (I-SCAcc), and also containing a repertory of tropes. The bibliography of the artcle is updated to the year 2016.

187 CANTUS PLANUS 2014 GABRIELLI — TROPES IN CANTUS PLANUS SOURCES FROM SOUTH TYROL

small church in the Badia valley, in Pieve di Marebbe/Plì de Mareo/Enneberg. Each Bressanone, Diocesan Museum gradual-kyriale 23 1493 archive also holds many liturgical fragments with musical notation. Table 1 lists the [I-BREmd], s.s. (Gradual of Taisten) medieval music manuscripts currently preserved in South Tyrol. Novacella, Augustinian Monastery 24 gradual-antiphoner 1495 [I-NV], Cod. 139 Table 1. Chant manuscripts now preserved in South Tyrol listed in chronological order (until ca. 1550) 25 Bressanone, Theological Library [I-BREs], A 17 gradual-kyriale end 15th century Sources with tropes: nos. 3, 7, 10-12, 14-16, 18-25 26 Bolzano, Franciscan Convent [I-BZf], CB.10 antiphoner 1519 / 17th century th Bressanone, Theological Library missal late 11 - first half 16th 1 th 27 Bressanone, Theological Library [I-BREs], F 6 pontifical [I-BREs], E 3 (Missal of Lavant) early 12 century century Bressanone, Theological Library missal first half 12th 2 28 Bressanone, Diocesan Archive [I-BREd], s.s. gradual-kyriale 15th- 16th century [I-BREs], B 22 (Missal of Karnol) century late 15th- th 29 Bolzano, Franciscan Convent [I-BZf], CB.1 processional 3 Bressanone, Vinzentinum [I-BRE], Inc. 6 missal 12 century early 16th century 4 Castle Tyrol, cod. 60 missal 12th century Bressanone, Theological Library late 12th - 5 missal [I-BREs], A 9 early 13th century State of the art 6 Castle Tyrol, cod. 61 missal late 13th - early 14th As regards the previous studies on these manuscripts, we should recall that a large number of them already appear in the catalogue compiled in 1905 by Hermann 7 Bolzano, City Museum [I-BZmc], ms. 1304 missal 1296 Julius Hermann, Die illuminierten Handschriften in Tirol; more recently, in the Iter Bressanone, Theological Library 8 antiphoner-hymnary 1372 [I-BREs], K 7 Liturgicum Italicum by Giacomo Baroffio (2011, pp. 61-62, 69-70, 324-325, 434-435, 289) and in the list compiled by Marco Gozzi for the volume Musikgeschichte Tirols 9 Bressanone, Capuchin Convent [I-BREc], s.s. gradual-kyriale 1426 (2001, pp. 543-550). Over time, some of the manuscripts have moved to other locations, 10 Bolzano, City Museum [I-BZmc], ms. 7/4 gradual-kyriale 1430 so their listings in the literature might no longer be completely accurate. For example, Novacella, Augustinian Monastery gradual-kyriale the two medieval manuscripts indicated by Hermann in 1905 as being located in the 11 1442 [I-NV], s.s. (Zollner Gradual 1) Franciscan convent of Bolzano are now held in the Franciscan provincial archive at

Novacella, Augustinian Monastery gradual-kyriale mid 15th century 12 Hall in Tirol, Austria. In the Franciscan archive in Bolzano, research has brought to [I-NV], s.s. (Zollner Gradual 2) (after 1442) light two other previously unknown medieval manuscripts with musical notation (nos. Bolzano, Benedictine Monastery Muri-Gries antiphoner 13 15th century 26 and 29 in the list). Also unknown was the gradual-kyriale, signed and dated 1426, [I-BZb], Cod. memb. 1 (Dominican) and held in the Capuchin Library at Bressanone (no. 9). There is, however, currently Novacella, Augustinian Monastery 14 antiphoner 15th century [I-NV], Cod. 15063 no trace of the two fifteenth century manuscripts from the cathedral of Bressanone Bressanone, Diocesan Museum that were held respectively in the Chapter Library and in the Diocesan Museum, 15 gradual-kyriale 15th/18th century [I-BREmd], s.s. now preserved on microfilm in the Bruno-Stäblein-Archiv in Würzburg (both with 2 16 Pieve di Marebbe, Parish Archive, s.s. gradual-kyriale 15th century the shelfmark I-BREd, S. n.) and mentioned in Troparia tardiva by Andreas Haug (1995, p. 26, nos. 012 and 013). The first one (no. 012) also appears in numerous Bressanone, Theological Library second half 17 pontifical [I-BREs], C 8 15th century publications from the 1950s on (Bosse 1955; Landwehr-Melnicki 1955; Thannabaur San Candido, Collegiate Church Museum second half 1962; Schildbach 1967; Stäblein 1975, pp. 202-203 with illustration; and Miazga 1976), 18 gradual-kyriale [I-SCAcc], VII a 7 15th century next to the gradual of Taisten (no. 23 in the list). The Bruno-Stäblein-Archiv also holds Bolzano, Benedictine Monastery Muri-Gries gradual-kyriale- 19 late 15th century on microfilm the gradual VII a 7 of San Candido (no. 18), containing numerous introit [I-BZb], Cod. m. 5 antiphoner tropes published in Troparia tardiva; in this case, the manuscript is still in its original Bressanone, Theological Library 20 gradual-kyriale 1476 location at the Museum of the Collegiate Church. Three fifteenth-century manuscripts [I-BREs], F 7 currently held in South Tyrol are mentioned in the volumes of Analecta Hymnica: the Bressanone, Theological Library 21 gradual-kyriale ca. 1480 [I-BREs], F 5 gradual-kyriale from Bolzano, City Museum 7/4 (no. 10), which contains some rare 22 Bolzano, City Museum [I-BZmc], ms. 7/3 gradual-kyriale ca. 1490 2. http://www.staeblein.musikwissenschaft.uni-wuerzburg.de/staeblein/search.php (last accessed in February 2015).

188 189 CANTUS PLANUS 2014 GABRIELLI — TROPES IN CANTUS PLANUS SOURCES FROM SOUTH TYROL sequences edited in volume 54,3 and two manuscripts now preserved in Bressanone: The manuscripts and their content the gradual of Taisten (no. 23) and ms. F 7 of the Theological Library (no. 20). The In terms of their general characteristics and content, the manuscripts, which with the gradual of Taisten is a large choirbook compiled in 1493 for the church of Taisten, a exception of the two Franciscan manuscripts and the Dominican codex were all written small village in the Pustertal valley situated between Bressanone and Lienz. It contains in Germanic adiastematic notation or in Hufnagelschrift, show a clear relationship about ninety sequences and many tropes. The manuscript has been on display for with the Germanic area and with the ecclesiastical province of Salzburg in particular. many years in the Diocesan Museum in Bressanone but as yet it has no number and, The codices kept at the Augustinian Abbey of Novacella, founded in 1142 and housing until the start of the present research, no modern foliation. The two volumes of the monks from Klosterneuburg, are linked to the manuscripts of the parent institution and famous Zollner Gradual (nos. 11 and 12) seem to have shared a similar fate: compiled those of the Augustinian abbey of Seckau. Tropes and sequences are present in great for the Augustinian abbey of Neustift by Friedrich Zollner in 1442 and subsequent numbers in these sources, just as in many medieval manuscripts from the Germanic years, they have always been known for their marvellous illuminations but have area. This notwithstanding, the first – and oldest – example of a trope discovered never been the object of any systematic musicological study.4 In South Tyrol are also in South Tyrol reveals a different origin. It is a fragment currently preserved in the preserved a significant number of manuscripts dating back to the twelfth century, monastery archive of Muri-Gries in Bolzano, a complete bifolium from a parchment listed as nos. 1-7. These belong to a type of manuscript that was fairly widespread in prosulary, measuring approximately 140 × 200 mm. The fragment probably survived the twelfth and thirteenth centuries, namely the Missale notatum (Praßl 2007, pp. 84- because of its reuse in the modern era, perhaps in the binding of a volume of smaller 85), a liturgical book comprising several different sections (usually calendar, missal, dimensions. The writing is well preserved on one side of the bifolium, while on the sacramentary, gradual, kyriale, prosar) assembled for small churches and chapels other the text is almost illegible due to heavy wear; only the rubrics have maintained and generally used by a single ministrant. Among these, codex 60 at Castle Tyrol has their original brilliance. The provenance of the fragment is unknown, but the musical undoubtedly been the most extensively studied. In the 1920s it was the subject of a notation allows us to conjecture a relation with a codex written in the Po valley area, master’s thesis discussed at the Pontifical Institute of Sacred Music (Eccher 1926); the possibly between Verona and Mantua. The fragment dates from the beginning of codex also appears in the publications of the monks of Solesmes (Le Graduel Romain, the eleventh century. The content and musical notation of the fragment show many 1957). It has also been widely discussed and investigated in a publication by Marco affinities in fact with the well-known Codice 107 (105) at the Biblioteca Capitolare in Gozzi devoted entirely to two music manuscripts kept at Castle Tyrol (2012, with a Verona, written in the Mantuan area. It shows substantial use of litterae significativae. full photographic reproduction of the manuscripts on the enclosed cd). This latter The notation shows particular affinities with fols. 51v-53 of the Veronese manuscript study forms part of a publishing project by the universities of Bolzano and Trento and its sequence for Saint Benedict, a piece added in another hand. The fragment devoted to the little researched or known printed and handwritten liturgical sources contains prosulae of the alleluia and offertory already present in the above-mentioned 5 held both in the historical region of Tyrol and elsewhere. On the three manuscripts Veronese codex and will soon be the subject of a specific publication. The oldest in Bolzano’s City Museum (nos. 7, 10 and 22) we should mention the extensive article ‘local’ manuscript with tropes is the calendar-gradual-kyriale-prosar-sacramentary by Marco Gozzi (2003) and the contribution by Jurij Snoj (2013) on the oldest of the preserved in Bressanone in the Vinzentinum private school, dating back to the twelfth three manuscripts (no. 7), copied in 1296 by a scribe from Ljubljana.6 With regard to century. The content of the source shows remarkable similarities with the missals of the most recent studies, we should also remember that the medieval manuscripts of the Salzburg area (see the list of alleluias for Sundays after Pentecost). We can suppose Novacella and of the Theological Library in Bressanone have in the last years been that the manuscript was in use in the diocese of Bressanone, but it was probably also subject to the first scientific cataloguing of medieval manuscripts in South Tyrol used in the diocese of Trent, as proper saints of this diocese are added to the calendar. using modern criteria, directed by Ursula Stampfer. A catalogue of the manuscripts The Saint Gallen-type notation includes the use of the Halbton-Torculus (Engels 1998; (including non-musical manuscripts) will soon be available in print. The description Praßl 1996, p. 22, and 2007, p. 93). The missal contains some tropes for the proprium of the music manuscripts (corresponding to nos. 1-2, 5, 8-9, 11-12, 14, 17, 21, 24-25, 27 missae, including the introit trope Hodie cantandus est for Christmas, as occurs in in the list) was drawn up with the collaboration of the present author. many sources from the Germanic area (Haug 1995, p. 10). The Tutilos trope is also contained in missal no. 7 dating from 1296. And it cannot be excluded that the trope 3. Gaude mater luminis (Ah 54 p. 358 no. 225), Ave sidus lux dierum (Ah 54 p. 387 no. 247), Gaude Maria templum (Ah 54 p. 333 no. 213). was also present in Codex no. 60 at Castle Tyrol (no. 4), which unfortunately contains

4. Engels (2001, pp. 309-310) provides a brief description of the musical notation of the two volumes and a transcription a serious lacuna regarding the Christmas formulary. of the Visitatio sepulchri from the Novacella gradual-antiphoner 139 (no. 24). 5. Daniele Torelli (2016) has recently published, in the same series, a study on the Franciscan antiphoner of Bolzano, A larger number of tropes are present in the later sources. The well-known gradual dating from 1519 and now preserved in Hall in Tirol. from San Candido contains ten introit tropes, edited by Andrea Haug (1995) in Troparia 6. The missals of Lavant and of Karnol (nos. 1 and 2 in the list) are described by Unterkircher (1979 and 1980).

190 191 CANTUS PLANUS 2014 GABRIELLI — TROPES IN CANTUS PLANUS SOURCES FROM SOUTH TYROL tardiva. The tropes from San Candido seem to be closely related to those attested in the it unnecessary to include either the full text of the pieces or their melody. Tropes with diocese of Freising, but also testify to the existence of an autonomous melodic tradition these text incipits appear neither in the bibliography nor in other sources, such as the (ibidem, p. 6, pp. 204-206, with transcription). A local tradition is attested in other manuscript in Innsbruck University Library A-Iu 457, a manuscript from Novacella fifteenth century manuscripts, which represent the largest part of South Tyrolean that contains a large number of tropes also in simple polyphony. The Innsbruck medieval sources. If we exclude the San Candido gradual, the South Tyrolean source source, however, contains another trope contained in one of the manuscripts now with the largest number of introit tropes is the above-mentioned Zollner gradual. It preserved in Novacella, antiphoner no. 14: the two-voice trope Procedentem sponsum contains the ‘traditional’ introit tropes for Christmas (Hodie cantandus est) with a de thalamo (a transcription of the melody of which appears in Stäblein 1956, p. remarkable illumination for the Puer natus, and the introit trope for Epiphany (Haec est 196). The piece is usually considered to be a trope of Benedicamus and appears in praeclara). The second volume of the gradual also contains traces of three other introit the Novacella antiphoner after the first vespers for Saint Stephen. In the Novacella tropes: two for Marian feasts and one for the patron Saint Augustine. Unfortunately, sources we can find an example of the same trope used in different liturgical-musical the manuscript only preserves the textual incipit of the tropes. In all three cases, a later contexts. The trope Mater Christi veneranda is generally used as a responsory trope hand (attributable to the end of the fifteenth century or the beginning of the sixteenth) or as a sequence (Ah 48 p. 428 no. 399; Hofmann-Brandt 1971, p. 80 no. 403). In fact, cancelled the trope text with a red line and added «non canitur» (or «non dicitur») on in the Novacella antiphoner it is related to the responsory O preclara stella maris, but a small area of scraped parchment.7 Here is the transcription of the text incipit of the in the Zollner gradual it is inserted in the offertory Filiae regum. Both pieces were tropes, marked with an asterisk (*): sung for the feast of Visitatio Mariae. Another offertory trope for Mary in the Zollner Gradual is the famous prosula Ab hac familia (Ah 49, p. 321 no. 624). This trope is In assumptione Bvm widespread in the German context and is also widely studied (Göllner 1985). We Gaudeamus also find it in the South Tyrolean manuscripts (nos. 10, 12, 19-20, 22-24), but what Quia virgo* is more characteristic is the constant presence of another offertory prosula to the Verbum verbo* same music, the trope Ave virgineum corpus Domini of the offertory Sacerdotes for Nunc Mariam* the feast of Corpus Christi. This trope appears in five manuscripts (nos. 19-21, 23-24) Nam se virgo* and seems to be a peculiar trait of South Tyrolean sources, hardly present in other areas. Analecta Hymnica gives just two references for this trope, one manuscript In nativitate Bvm from Bressanone and one from Admont (Ah 4, p. 325 no. 643). The fact that it also Gaudeamus appears as in addition in manuscripts from Bolzano is also interesting, as the latter Nata summa* area actually belonged to the diocese of Trent. Verbum verbo* Nunc Mariam* A further aspect emerging from analysis of the repertoire is the presence of a large Nam se virgo* number of ordinarium missae tropes. The Zollner gradual features a variety of these, but other manuscripts do as well. In some cases, these seem to be somewhat ‘rare’ In festo sancti Augustini tropes. One example is a trope of the Agnus from the Zollner gradual, whose the Statuit ei melody is signalled in Schildbach (1967, p. 122 no. 149), but lacks any reference to Augustino* tropes.8 The text is as follows: Hic est manu* Ut pro nobis* Agnus Dei qui tollis peccata mundi. Hic est cleri* O dulcis Ihesu Christe ab arce patris misse We can assume that these compositions had a rhythmic text (trochee). Trope and tu redemisti omnia introit were sung alternatim by chorus and cantores, as indicated in the rubrics. It can nostra delesti crimina. reasonably be assumed that the tropes were sung from memory, as the copyist deemed Dona nobis pacem.

7. A similar case is reported by Haug (1995, p. 10) in reference to a gradual from the Minden cathedral where, at the beginning of the sixteenth century, a later hand added in the margins: Non servatur. 8. My heartfelt thanks go to Charles Atkinson for his help in research around this trope.

192 193 CANTUS PLANUS 2014 GABRIELLI — TROPES IN CANTUS PLANUS SOURCES FROM SOUTH TYROL

Another example is the alternative text to the Kyrie Fons bonitatis for the dedication Bibliography of a church from manuscript F 7 (no. 20) preserved in the Theological Library of Bressanone, which in turn originates from near the city of Bressanone. Here is the text: Baroffio Giacomo (2011), Iter liturgicum italicum. Editio maior, Stroncone: Associazione San Michele Arcangelo.

Kyrie Bosse Detlef (1955), Untersuchung einstimmiger mittelalterlicher Melodien zum “Gloria in excelsis”, sacrator summe templorum Domine: Regensburg: Bosse (Forschungsbeiträge zur Musikwissenschaft 2).

domum istam benedic pie Cantus: a Database for Latin Ecclesiastical Chant eleison. http://cantus.uwaterloo.ca (last accessed in February 2015). Christe Dreves Guido M. - Blume Clemens - Bannister Henry M. (1886-1922), Analecta hymnica Medii Aevi [Ah], celice sit hec aula curie consors Leipzig: Reisland. in tuo dedicato sacrato nomine: Eccher Celestino (1926), Il codice 60 di Castel Tirolo. Illustrazione liturgico-musicale con accenni al luogo quam de celo clemens respice di provenienza ed all´opera di composizione dello stesso, Roma: Pontificia Scuola Superiore di Musica eleison. Sacra, a.a. 1926 (unpublished dissertation).

Engels Stefan (1998), ‘Adiastematische Neumen mit melodischer Zusatzbedeutung. Ein wichtiges Hilfsmittel Kyrie zur Melodierestitution’, Beiträge zur Gregorianik 26, pp. 63-80. audi de celis precum voces – (2001), Einstimmige liturgische Handschriften des Mittelalters in Tirol in Kurt Drexel - Monika quas fidelis effundit populous: Fink eds., Musikgeschichte Tirols, 3 vols., Innsbruck: Wagner (Schlern-Schriften 315), vol. 1: Von den hoc in templo sit propicius Anfängen bis zur frühen Neuzeit, pp. 219-332. eleison. Haug Andreas (1995), Troparia tardiva. Repertorium später Tropenquellen aus dem deutschsprachigen Raum, Kassel: Bärenreiter (Mmmae Subsidia 1).

Hermann Hermann Julius (1905), Die illuminierten Handschriften in Tirol, Leipzig: Hiersemann. Finally, I would also like to underline the presence of tropes with rhythmical notation. One example is the above-mentioned Agnus, although the stems seem to have been Hofmann-Brandt Helma (1971), Die Tropen zu den Responsorien des Officiums, 2 vols., Ph. D. dissertation, added later. Another, decisively rhythmical example is the Gloria Spiritus et alme Friedrich-Alexander Universität, Erlangen-Nürnberg

(Bosse 1955, p. 99 no. 49), present in two manuscripts (nos. 10 and 24), with its use Gabrielli Giulia (2015), I manoscritti liturgico-musicali della città di Bolzano, Lucca: Lim (Bibliotheca of semibreves and minimae (in the gradual-antiphoner from Novacella, the minimae Mediaevalis 3). 9 have red stems ). Finally, the alleluia trope O Maria caeli via (Cantus no. 507047) has Gozzi Marco (2001), Musikgeschichte der Region Trient bis 1600, in Kurt Drexel - Monika Fink eds., a ternary-rhythmical version in two South Tyrolean manuscripts (nos. 18 and 20 from Musikgeschichte Tirols, 3 vols., Innsbruck: Wagner (Schlern-Schriften 315), vol. 1: Von den Anfängen San Candido), notated using two, closely written puncta to indicate a longer sound. bis zur frühen Neuzeit, pp. 467-594. The tropes in cantus planus sources from South Tyrol provide us with a relatively – (2003), ‘Tre sconosciuti manoscritti liturgici con notazione musicale del Museo Civico di Bolzano, vast and interesting repertoire that shows, on the one hand, the characteristics of late Studi trentini di scienze storiche 82/4, pp. 729-777. sources from the Germanic area, and on the other, its own unique characteristics that – (2012), I codici liturgici di Castel Tirolo, Lucca: Lim (Monumenta Liturgiae et Cantus 1). are deserving of further analysis and in-depth research. Göllner Marie Louise (1985), Musical setting of the trope «Ab hac familia» in Peter Dronke ed., Liturgische Tropen: Referate zweier Colloquien des Corpus Troporum in München (1983) und Canterbury (1984), München: Arbeo, pp. 89-106.

Le Graduel Romain. Édition critique par les moines de Solesmes (1957), Solesmes: Abbaye Saint-Pierre, II: Les sources.

Landwehr-Melnicki Margaretha (1955), Das einstimmige Kyrie des lateinischen Mittelalters, Regensburg: Bosse (Forschungsbeiträge zur Musikwissenschaft 1).

Miazga Tadeusz (1976), Die Melodien des einstimmigen Credo der römisch-katholischen lateinischen Kirche. Eine Untersuchung der Melodien in den handschriftlichen Überlieferungen mit besonderer 9. An illustration is in Engels (2001, p. 309). Berücksichtigung der polnischen Handschriften, Graz: Adeva.

194 195 CANTUS PLANUS 2014

Marit Johanne Høye Praßl Franz Karl (1996), ‘Choralhandschriften österreichischer Augustinerchorherren im 12. Jahrhundert’, Musicologia Austriaca 14/15, pp. 9-32. Kyrie chants in manuscripts – (2007), Codex Wien Österreichische Nationalbibliothek 13314: Das älteste Klosterneuburger Graduale (Missale)? in Birgit Lodes ed., Wiener Quellen der älteren Musikgeschichte zum Sprechen gebracht, from the German-speaking area Tutzing: Schneider (Wiener Forum für ältere Musikgeschichte 1), pp. 83-109.

Schildbach Martin (1967), Das einstimmige Agnus Dei und seine handschriftliche Überlieferung vom 10. bis zum 16. Jahrhundert, Erlangen: Hogl.

Snoj Jurij (2012), Ruotlib’s notated missal and its historical background in Robert Klugseder ed., Cantus Kyrie chants are known to have been transmitted in a variety of different melodic th Planus: papers read at the 16 meeting, Vienna, Austria, 2011, Purkersdorf: Hollinek, pp. 361-366. forms. The earliest surviving manuscripts transmitting notated Kyrie chants are from

Stäblein Bruno (1956), Hymnen I. Die mittelalterlichen Hymnenmelodien des Abendlandes, Kassel: the first half of the tenth century: the Aquitanian F-Pn 1240 of Saint Martial, from the Bärenreiter (Mmmae 1). west-Frankish area, and two manuscripts from Saint Gall, CH-SGs 484 and CH-SGs

– (1975), Schriftbild der einstimmigen Musik, Leipzig: Veb Deutscher Verlag für Musik (Musikgeschichte 381, from the east-Frankish area. Saint Martial and Saint Gall were both important in Bildern III/4). liturgical centres at the time. Previous research on the Kyrie chant has found significant differences between early east-Frankish and west-Frankish repertories, as notated in Thannabaur Peter Josef (1962), Das einstimmige Sanctus der römischen Messe in der handschriftlichen Überlieferung des 11.-16. Jahrhunderts, München: Ricke. Saint Gall and at Saint Martial before the mid-tenth century. Characteristic of early melodies notated in Saint Gall is their formal representation in simple repeated Torelli Daniele (2016), L’Antifonario del convento dei Francescani di Bolzano, Lucca: lim (Monumenta Liturgiae et cantus 3). patterns ii or iii, e.g., pattern aaa bbb aaa or aaa bbb ccc, short melodic lines and their association only with tropes. Characteristic of early Aquitanian melodies is their formal Unterkircher Franz (1979), ‘Ein altes Missale von Lavant’, Osttiroler Heimatblätter 47/12. representation in more complex repeated patterns, in particular pattern iv, e.g., axa – (1980), ‘Das Missale von Karnol’, Der Schlern 54, pp. 129-139. byb czc, extended melodic lines and the Latin-texted form (Høye 2008, 1, p. 227). Kyrie chants with melodic pattern axa byb czc, a pattern now believed to have originated in Aquitaine, were transmitted to northern France, where different melodic versions of a chant coexisted. But what happened to Kyrie melodies in the east-Frankish area? Latin texts and tropes used with Kyrie melodies were often changed when the chant was adopted in a new region. This study asks whether such a regional change is also reflected in the melodic outlining of a Kyrie chant. In the following, a select group of Kyrie chants that was disseminated in both Anglo-French and German-speaking areas, will be examined. Kyrie 47, Kyrie 155 and Kyrie 55 belong to the earliest layer of Kyrie chants and were already notated in the first half of the tenth century. The three chants will be examined in their earliest transmissions, in the neumatic notation of the tenth- and eleventh-century manuscripts. Two other Kyrie chants, Kyrie 68 and Kyrie 58, will be examined in later manuscripts with diastematic notation. The focus of this study will be on identifying formal shapes and regional melodic variants in order to shed further light on the varied melodic identity of the chants in the Kyrie repertory.

Kyrie 47, a chant first notated at Saint Martial in the west-Frankish area Kyrie 47 was first notated in the Aquitanian manuscript F-Pn 1240 with melodic repeated pattern iv, axa byb czc’. The chant achieved wide dissemination in the west- Frankish area. Kyrie 47 was transmitted in the east-Frankish area as well, but it was never widely disseminated. Margareta Melnicki (1955, p. 94) gives only five German

196 197 CANTUS PLANUS 2014 HØYE — KYRIE CHANTS IN MANUSCRIPTS FROM THE GERMAN-SPEAKING AREA manuscript concordances for the chant together with one concordance from the Czech The early adiastematic tenth- and eleventh-century manuscripts do, however, give Republic, and I have found one more: the north-west German D-KNd 1001b from more concordances for Kyrie 47; four Rhenish and three German manuscripts transmit Cologne. For an example of Kyrie 47 with repeated pattern iv from a later German this chant (Bjork 1979/80, pp. 13, 34). An examination of these early manuscripts manuscript, see example 1. shows that Kyrie 47 maintained its melodic properties when the chant was transmitted into German-speaking areas. Already in the two tenth-century Rhenish manuscripts Example 1. Kyrie 47 with melodic repeated pattern iv, axa byb czc’, D-KNd 1001b, fol. 306 GB-Lbl 19768 and F-Pn 9448, the chant was transmitted with melodic repeated pattern iv. The axa byb czc’ pattern was also used with the chant in later German and Rhenish sources. Kyrie 47 never gained widespread use in these regions, however, and remained a typical ‘western’ Kyrie chant.

Kyrie 155, a chant first notated in Saint Gall in the east-Frankish area The earliest notation of Kyrie 155 is in Saint Gall manuscripts CH-SGs 484 and CH-SGs 381, from approximately the same time as Kyrie 47 was notated in Aquitaine. In the two Saint Gall codices, all four Kyrie melodies in the main body of the manuscripts were notated with melodic repeated patterns ii, aaa bbb aaa, and III, aaa bbb ccc. Kyrie 155 is notated with pattern iii in these codices (see CH-SGs 381, p. 295). For an example of Kyrie 155 with repeated pattern iii from a later German manuscript, see example 2.

Example 2. Kyrie 155 with melodic repeated pattern iii, aaa bbb ccd, A-Gu 9, fol. 165v

Kyrie 155 was transmitted in the west-Frankish area as well, and the chant eventually became widespread in both areas. Kyrie 155 was notated in all surviving Aquitanian manuscripts from the end of the tenth century (Bjork 1979/1980, pp. 13-14). In line with the general melodic transmission of Kyrie melodies in Aquitaine at the time, Kyrie 155 was notated here with the axa byb czc-form of pattern iv. The chant is also notated

198 199 CANTUS PLANUS 2014 HØYE — KYRIE CHANTS IN MANUSCRIPTS FROM THE GERMAN-SPEAKING AREA with the same complex repeated pattern in the English GB-Ccc 473 from the late tenth 2013, pp. 839‑854).The chant has nevertheless been viewed as unusually stable in its century. We already know that the more complex repeated pattern had reached both melodic form. It is therefore of no surprise to find that Kyrie 55, like Kyrie 47 discussed Rhenish manuscripts and those in Saint Gall by the second half of the tenth century, above, was notated in German manuscripts, both diastematically and adiastematically, through transmission of Kyrie 55. Kyrie 55 was included in the main body of the with the complex repeated pattern first used in the Aquitaine. Kyrie 55 in CH-SGs Rhenish GB-Lbl 19768, and the chant was added to the two Saint Gall codices during 484 and CH-SGs 381, which was added in the second half of the tenth century, was the second half of the tenth century (Bjork 1979/80, p. 13). All three entries of Kyrie 55 already notated with the complex melodic pattern iv, although only the Greek text in these manuscripts appear with melodic repeated pattern iv, a melodic pattern that is appears, not the Latin. In the north-French transmission, I found one exception to often considered as a defining feature of this particular Kyrie chant. Kyrie 155 is notated this otherwise quite universal transmission of Kyrie 55 with repeated pattern iv. in the two Saint Gall codices with melodic pattern iii, aaa bbb ccd. The chant is notated The Nevers manuscript F-Pn 1235 gives two transmissions of Kyrie 55, one with the with the same pattern in a German source that is only slightly more recent: the late standard complex pattern axa byb czc’ and one with a mixed melodic pattern where tenth-century D-BAs 5 from nearby Reichenau. Moreover, a late tenth-century source only the last three phrases are contrasted: aaa bbb cxc’. This would seem like a unique from Heidenheim, GB-Ob 27, also transmits repeated pattern iii with this melody. All version of Kyrie 55. When examining more of the Saint Gall codices, however, two of of these sources are located in or near southern modern-day Germany. Further north, the later manuscripts CH-SGs 376 and CH-SGs 380 from the second half of the eleventh however, we find a different melodic transmission when another late tenth-century century transmit the same mixed melodic repeated pattern: aaa bbb cxc’ (see figure 1, source, the Rhenish F-Pn 9448 from Prüm, transmits Kyrie 155 with the complex p. 202-203 for Kyrie 55 in CH-SGs 380). The other Saint Gall manuscripts that transmit repeated pattern iv, known from the Aquitaine. The two slightly younger eleventh- Kyrie 55 (CH-SGs 484, CH-SGs 381, CH-SGs 378 and CH-SGs 546) all comply with the century sources D-Mbs 14083 and D-Mbs 14322 from Saint Emmeram, Regensburg, standard transmission of this melody, with melodic repeated pattern axa byb czc’. also transmit this version of the melody. Furthermore, several later German sources with diastematic notation have been found to transmit this chant with repeated pattern A partly notated Kyrie chant in a Saint Gall manuscript iv (see, for instance, D-LEu 391, fol. 117v and D-KEsm 9, fol. 259). These findings I have previously established that the variations between what is perceived as different suggest that although the simpler melodic pattern iii might have been the first version melodic forms, or repeated patterns, with a Kyrie chant, are often not extensive when to circulate in the east-Frankish area with Kyrie 155, at least in the more southern found for the same melody. Most often, only the opening of a contrasting, middle parts, the more complex melodic version nevertheless became equally favoured as phrase is varied from its surrounding phrases, as we can see in Kyrie 47 in example 1. early as in the eleventh century, as witnessed by D-Mbs 14083 and D-Mbs 14322. The Furthermore, the nature of these variations suggests that more varied forms may have two melodic repeated patterns transmitted with Kyrie 155 seem to have coexisted in originated as more repetitive melodic outlines. Thus, the complex melodic pattern iv the east-Frankish area, in the same way that they did in the north-French transmission with a contrasting middle phrase is more likely to have been derived from the simpler (Høye 2008, 2, p. 59). In Saint Gall, however, Kyrie 155 was notated with repeated patterns ii or iii (Høye 2008, 1, pp. 96-123 and 230). Interestingly, an incomplete Kyrie pattern iii also in the second half of the eleventh century, as witnessed by CH-SGs 378 chant notated in manuscript CH-SGs 378 displays a notation technique that might in and CH-SGs 380. In addition, in CH-SGs 378, a second notated version of Kyrie 155 that fact support the notion of such a procedure with the composition of Kyrie melodies. was added as late as in the thirteenth century still contains this simpler pattern. Thus, For some reason, the scribe never finished notating this Kyrie. This provides us with a it would seem like the scribes in Saint Gall may have had a preference for the simpler unique opportunity for analysing how he set out to notate the first three phrases of the melodic repeated pattern with Kyrie 155. chant (see p. 375 of CH-SGs 378 in figure 2, p. 209).

Kyrie 55, a widespread chant first notated in the west-Frankish area Phrases 1 and 3 are notated in full, and their melodic lines are identical. In the second Kyrie 55 is commonly found in tenth-century Kyrie-manuscripts, and the chant has phrase, however, the scribe notated neumes only for the last part of the phrase, over received much attention due to several aspects of its melody. Kyrie 55 was the first ‘eleison’. The notated part is identical to the surrounding phrases 1 and 3. But empty Kyrie notated in the Aquitanian F-Pn 1240, a manuscript notated several decades space has been left for the opening of phrase 2, over the acclamation ‘Kyrie’. An empty before other surviving Aquitanian manuscripts. Further, the chant was set with a full space in this position indicates that the scribe intended to add a different melodic Latin text as well as being notated with the complex melodic repeated pattern iv. The motive to it later, perhaps a melodic motive that was not yet known to him or that he chant’s inclusion in the early F-Pn 1240 has been used to argue that Kyrie 55 shows did not remember. The addition of a new melodic motive in the opening of phrase 2 the earliest form of the Kyrie melodies, signified by the Latin text and the complex would transform the shape of the melody into the complex melodic pattern iv, axa byb axa byb czc-pattern (Bjork 1981, p. 16). This view has later been differentiated (Høye czc, with a contrasting second phrase.

200 201 CANTUS PLANUS 2014

Figure 1 Kyrie 55 in Saint Gall, Stiftsbibliothek, Cod. Sang. 380, pp. 112-113 Calendario, Computo, Tropario, Sequenziario HTTPS://WWW.E-CODICES.CH/IT/LIST/ONE/CSG/0380

202 203 CANTUS PLANUS 2014 HØYE — KYRIE CHANTS IN MANUSCRIPTS FROM THE GERMAN-SPEAKING AREA

Melodic variation in the transmission of Kyrie 68

Example 3a. Melodic variations transmitted with Kyrie 68, phrase 1 Example 3b. Melodic variations transmitted with Kyrie 68, phrase 4

Var. 1 Var. 2 Var. 1 Var. 2

F-Pn 10508 (St Evroult) F-Pn 10508 (St Evroult) (west) (west) œœ œœ œœ & œœœœœœ œ œ œœœœ œ œ & œ œ œ œ œ œœœ œœœœ œ œœ œ Ky - riœ - eœ œ œ œ œ œ œ œ leiœ - son. Chri - ste œ œ œ œ lei - son.

D-Mu 2 156 (Moosburg) D-Mu 2 156 (Moosburg) (east) (east) œ œœ œœ & œœœœ œ œœ œ œœœœœœœ œ œ & œ œ œ œ œœ œœœœœ œ œ Ky - riœ œ œ - eœ œ œ œ œ œ leœ - y -son. Chri - ste œ œ œ œ leyœœ - son.

Variation 1 Variation 1

west west E-Mn 289, F-Pn 10508, RUS-SPsc OvI6, F-AN 96, F-AN 97, F-PR 12, I-Ac 695, F-Pn E-Mn 289, F-Pn 10508, RUS-SPsc OvI6, F-AN 96, F-AN 97, F-PR 12, I-Ac 695, F-Pn 1112, F-Pn 14452, F-RS 264, F-LA 263, F-Pn 10511, F-Pn 1235, F-CA 61 (Latin texted 1112, F-Pn 14452, F-RS 264, F-LA 263, F-Pn 10511, F-Pn 1235, F-CA 61 (melismatic), and melismatic), F-VN 130, F-VN 98, F-Pn 833, F-CO 232, D-KNd 156, D-KNd 1001b, F-VN 130, F-VN 98, F-Pn 833, F-CO 232, D-KNd 156, D-KNd 1001b, CH-BSPh 3 D-Mbs 5539, CH-SGs 546 (Latin texted) east east F-CA 61 (Latin texted), CH-SGs 383, CH-SGs 546 (Latin texted and melismatic), CH-BSPh 3, CH-SGs 383, CH-SGs 546 (melismatic), D-KEsm 10, D-Mbs 19267, D-KEsm 10, D-Mbs 19267, D-Mbs 5539, D-Mbs 11764, D-Mu 2 156, D-LEu 391 D-Mbs 11764, D-Mu 2 156, D-LEu 391 Variation 2 Variation 2 west west E-Mn 289, F-Pn 10508, RUS-SPsc OvI6, F-AN 96, F-AN 97, F-PR 12, I-Ac 695, F-Pn E-Mn 289, F-Pn 10508, RUS-SPsc OvI6, F-AN 96, F-AN 97, F-PR 12, I-Ac 695, F-Pn 1112, F-Pn 14452, F-RS 264, F-LA 263, F-Pn 10511, F-Pn 1235, F-CA 61 (Latin 1112, F-Pn 14452, F-RS 264, F-LA 263, F-Pn 10511, F-Pn 1235, F-CA 61 (Latin texted), F-VN 98, F-CO 232, D-KNd 156, D-KNd 1001b, CH-SGs 546 (Latin texted) texted), F-VN 98, F-CO 232, D-KNd 156, D-KNd 1001b, CH-SGs 546 (Latin texted) east east F-CA 61(Mel), F-VN 130, F-Pn 833, D-Mbs 5539, CH-BSPh 3, CH-SGs 383, CH-SGs F-CA 61 (melismatic), F-VN 130, F-Pn 833, D-Mbs 5539, CH-BSPh 3, CH-SGs 383, 546 (melismatic), D-KEsm 10, D-Mbs 19267, D-Mbs 11764, D-Mu 2 156, D-LEu 391 CH-SGs 546 (melismatic), D-KEsm 10, D-Mbs 19267, D-Mbs 11764, D-Mu 2 156, D-LEu 391 Variation 1 = west + Variation 2 = east F-CA 61 (melismatic), F-VN 130, F-Pn 833, CH-BSPh 3 Variation 1 = west + Variation 2 = east F-CA 61 (melismatic), F-VN 130, F-Pn 833, D-Mbs 5539 Variation 1 = east + Variation 2 = west F-CA 61 (Latin texted), CH-SGs 546 (Latin texted)

204 205 CANTUS PLANUS 2014 HØYE — KYRIE CHANTS IN MANUSCRIPTS FROM THE GERMAN-SPEAKING AREA

Kyrie 68 differs somewhat from the three melodies I have discussed so far. While the CH-BSPh 3, and D-Mbs 5539 from Diessen in Germany. Thus, with Kyrie 68, several chant was notated with both melodic repeated pattern iii and iv, the varied melodic manuscripts transmit a blending of regional melodic features, and to a varying extent. pattern iv seems to only have been in use in Aquitaine. All but three Aquitanian While the two north-west German sources D-KNd 156 and D-KNd 1001b follow the manuscripts transmit the melody with melodic pattern iv (Høye 2008, 1, p. 135; Bjork Anglo-French tradition with all four melodic motives, the manuscript from further 1977, 2, pp. 175-202). In northern France, England and also in manuscripts from the south, Diessen, only transmits one Anglo-French melodic feature: variation 1 of phrase German-speaking area, however, the melody of Kyrie 68 is notated with the simpler 1. We see that the most frequent use of regional variations is found in manuscripts repeated pattern iii. I have therefore used a different approach toward this melody, from the eastern part of modern-day France, north-west Germany and, to some extent, examining Kyrie 68 in later, diastematic manuscripts from Anglo-French and in the Swiss manuscript CH-SGs 546. The latter finding was somewhat unexpected, German-speaking areas. Previous study of a group of sequence melodies in the east- since with other melodies, CH-SGs 546 tends to transmit a stable German melodic French manuscript F-Pn 833 showed that sequences in this manuscript transmit a tradition. However, the manuscripts F-CA 61 and CH-SGs 546 transmit the same Latin blending of regional melodic motives that are otherwise particular to Anglo-French text with Kyrie 68: Rex pie rex regum. The text has a limited transmission and has so areas or to German areas, respectively. In this, the manuscript to some extent far only been found in CH-SGs 546 and in three more north-east French manuscripts, adheres to other manuscripts in this area of modern-day France, bordering on the all three with adiastematic notation (F-CA 60 and F-CA 78 from Cambrai and F-CA Low Countries and Germany. Further examination showed variance in whether an 75 from Arras). Both F-CA 61 and CH-SGs 546 give a different melodic version for individual sequence melody leans more toward the German or the Anglo-French their Latin-texted phrases than for their melismatic ones (see examples 3a and 3b). tradition, even within the same source. Interestingly, we see under ‘Variation 2’ in both phrases 1 and 4, that each of the two manuscripts is included in the ‘wrong’ group when they correspond exactly to the I have examined Kyrie 68 as transmitted in F-Pn 833 to see to what extent such a other. The Latin-texted phrases 1 and 4 of both F-CA 61 and CH-SGs 546 transmit the blending of melodic traditions also applies to the Kyrie chants in this manuscript. western or Anglo-French version of the melody at this place, while the melismatic Examination of Kyrie 68 in twenty-nine manuscripts revealed regional features phrases 1 and 4 notated in the same manuscripts, transmit the German version of the reflecting a German and an Anglo-French tradition, and some of the results are melody. This would suggest that the melismatic and the Latin-texted Kyrie phrases summarised in examples 3a and 3b. Example 3a shows melodic variations in phrase might have come from two different melody transmissions. 1 of Kyrie 68, and example 3b shows melodic variations in phrase 4. The north-French manuscript F-Pn 10508 shows melodic features of an Anglo-French transmission, Melodic variation in the transmission of Kyrie 58 while the Moosburg graduale represents the melodic tradition transmitted in the As a final point, I would like to draw attention to the transmission of Kyrie 58. This German-speaking area. chant displays a melodic feature that is particularly interesting with regard to the present discussion. Kyrie 58 is transmitted with two different melodic patterns, I have identified two points of variance for each of the two phrases, and these are repeated pattern ii, aaa bbb aaa, and iii, aaa bbb ccc. For this chant, however, the marked with boxes. Under the headings ‘Variation 1’ and ‘Variation 2’, I have notated distribution of the different patterns does not separate an Anglo-French transmission each manuscript that transmits an Anglo-French or a German version, respectively. from a German transmission. In table 1 we see instead that Kyrie 58 is notated with German or Swiss manuscripts that transmit an Anglo-French melodic feature are a separate melodic transmission in manuscripts from eastern France; Laon, Reims given in italics, as are French manuscripts that transmit a German melodic feature. and Verdun, together with south-west Swiss sources. Thus, we see how manuscripts When a manuscript adheres to both traditions, by transmitting one Anglo-French from areas that transmit a blending of Anglo-French and German melody traditions and one German melodic motive in the same phrase, the manuscript is listed with Kyrie 68, instead give a melodic transmission distinct from both main traditions, accordingly under ‘Variation 1 = west + Variation 2 = east’. with Kyrie 58.

Examples 3a (p. 204) and 3b (p. 205) show how F-Pn 833 together with two more French manuscripts, F-CA 61 and F-VN 130, transmit parts of Kyrie 68 with German melodic features. The manuscripts are from the northeastern part of France: Lille, Verdun and Saint Stephen at Arne. Further, five German and Swiss manuscripts transmit parts of Kyrie 68 with Anglo-French melodic features: two north-west German sources from Wesel and Köln, D-KNd 156 and D-KNd 1001b, two Swiss sources, CH-SGs 546 and

206 207 CANTUS PLANUS 2014 HØYE — KYRIE CHANTS IN MANUSCRIPTS FROM THE GERMAN-SPEAKING AREA

Table 1. Regional variation in the melodic outlining of Kyrie 58

Siglum Provenance Melodic repeated pattern

F-CA 60 Cambrai aaa bbb aaa’

F-CA 78 Cambrai aaa bbb aaa’

F-DOU 90 Anchin aaa bbb aaa’

F-Pn 3126 Nevers axa byb ax’a’

F-Pn 1112 Paris aaa bbb aaa’

F-Pn 14452 Paris aaa bbb aaa’

F-Pn 1139 Aquitaine aaa bbb aaa’

F-Pn 1119 Aquitaine aaa bbb aaa’

F-Pn 1132 Aquitaine aaa bbb aaa’

F-LA 263 Laon aaa bbb ccc’

F-RS 264 Reims aaa bbb ccc’

F-RS 266 Reims aaa bbb ccc’

F-VN 759 Verdun aaa bbb ccc

F-VN 98 Metz? aaa bbb ccc’

F-Pn 833 Saint Stephen at Arne aaa bbb ccc’

CH-SGs 383 Lausanne? aaa bbb ccc

CH-BSPh 3 Bourg Saint Pierre aaa bbb ccc’

D-KNd 1001b Köln aaa bbb aaa’

D-KNd 156 Wesel aaa bbb aaa’

D-Aab 22 Augsburg aaa bbb aaa’

D-Mu 2 156 Moosburg aaa bbb aaa’

CH-SGs 546 Saint Gallen aaa bbb aaa’

D-Mbs 5539 Diessen aaa bbb aaa’

Figure 2 Incomplete Kyrie chant in Saint Gall, Stiftsbibliothek, Cod. Sang. 378, p. 375 Calendario, Computo, Tropario, Sequenziario HTTPS://WWW.E-CODICES.CH/IT/LIST/ONE/CSG/0378

208 209 CANTUS PLANUS 2014 HØYE — KYRIE CHANTS IN MANUSCRIPTS FROM THE GERMAN-SPEAKING AREA

Conclusion Appendix Summary of manuscript sources of Kyrie 68 An examination of early manuscript entries with Kyrie 47 and Kyrie 155 shows that the west-Frankish Kyrie 47 retained its melodic shape of the complex melodic pattern iv when travelling into new regions in the tenth century. The east-Frankish Kyrie 155 Siglum Provenance Date with melodic repeated pattern iii, on the other hand, was at some point redacted into CH-BSPh 3 Bourg Saint Pierre fourteenth century a more complex form. This is more likely to have happened in the west-Frankish area, CH-SGs 383 Lausanne? thirteenth century perhaps in Aquitaine. The new version with a more complex pattern was then re- introduced into German areas, perhaps via Rhenish sources, and remained in use CH-SGs 546 Saint Gall sixteenth century there, coexisting with its simpler version. The otherwise quite uniformly transmitted D-KEsm 10 Kerpen fifteenth century melody of Kyrie 55, with repeated pattern iv, shows another way in which a Kyrie chant D-KNd 156 Wesel sixteenth century could be modified according to local preference. The melody would seem to have been deliberately simplified, conveying a mixed melodic repeated pattern in Saint Gall and D-KNd 1001b Köln thirteenth century also at Nevers. Analyses of melodic variants with Kyrie 68 have shown how several D-LEu 391 Leipzig fourteenth century manuscripts transmit a blending of Anglo-French and German melodic features with D-Mbs 5539 Diessen fourteenth century this Kyrie. Analyses of Kyrie 58, on the other hand, show how manuscripts from the D-Mbs 11764 Polling fourteenth century same area give a melody transmission that is distinct from Anglo-French and German traditions. Both Kyrie chants show, however, that the most frequent use of such regional D-Mbs 19267 Regensburg fourteenth century melody variations is found in manuscripts from eastern France, north-west Germany D-Mu 20 156 Moosburg fourteenth century and, to some extent, south-west Switzerland. E-Mn 289 Palermo twelfth century

F-AN 96 Angers twelfth century

F-AN 97 Angers twelfth century

F-CA 61 Lille eleventh / twelfth century

F-CO 232 (Dominican) fourteenth century

F-LA 263 Laon twelfth/thirteenth century

F-Pn 833 Saint Stephen at Arne twelfth century

F-Pn 1112 Paris thirteenth century

F-Pn 1235 Nevers twelfth century

F-Pn 10508 Normandy twelfth century

F-Pn 10511 Auxerre thirteenth century

F-Pn 14452 Paris thirteenth century

F-PR 12 Chartres thirteenth century

F-RS 264 Reims thirteenth century

F-VN 98 Metz? fourteenth century

F-VN 130 Verdun fourteenth century

I-Ac 695 Reims thirteenth century

RUS-SPsc OvI6 Meulan twelfth century

210 211 CANTUS PLANUS 2014

Sarah Ann Long Bibliography

Bjork David (1977), The Kyrie repertory in Aquitanian manuscripts of the tenth and eleventh centuries, 2 vols., Mensurally notated mass ordinaries Ph. D. dissertation, University of California, Berkeley. from Cambrai and Tournai – (1979-1980), ‘Early repertories of the Kyrie eleison’, Kirchenmusikalisches Jahrbuch 63-64, pp. 9-43.

– (1981), ‘The Early Frankish Kyrie Text’, Viator 12, pp. 9-35.

Høye Marit Johanne (2008), The North French Kyrie Chant. A Historical and Stylistic Study of the Repertory before 1200, 2 vols., Ph. D. dissertation, Ntnu: Trondheim (http://urn.kb.se/resolve?urn=urn:nbn:no: Scholars of chant and polyphony have long been interested in musical traditions at the ntnu:diva-5013). cathedrals of Tournai and Cambrai in the fourteenth and fifteenth centuries. As the

– (2013), Two Early Kyrie Chants in Barbara Hagg-Huglo - Debra S. Lacoste - Nicolas Bell eds., seats of two neighboring dioceses, both institutions were important centers of Franco- International Musicological Society Study Goup Cantus Planus. Papers read at the 15th Meeting, Flemish compositional activity. Music at Cambrai cathedral has been widely studied, Dobogókö/Hungary 23-29 August 2009, 3 vols., Lions Bay: The Institute of Mediaeval Music, 1, pp. 839- 854. with composers such as Dufay, Tinctoris, and Ockeghem among its ranks, and the Médiathèque Municipale in Cambrai preserves many important archival collections Melnicki Margareta (1955), Das einstimmige Kyrie des lateinischen Mittelalters, Regensburg: Bosse and manuscripts documenting its vibrant artistic heritage (a full list is in Molinier (Forschungsbeiträge zur Musikwissenschaft 1). 1891). Music and compositional practice at Tournai cathedral is less well documented, although what can be pieced together suggests a comparably rich musical tradition. One of the first polyphonic mass ordinary compilations of the fourteenth century, known as the ‘Tournai Mass’ (Coussemaker 1862, pp. 100-110; Dumoulin et al. 1988, pp. 25-38; Guletsky 2009, pp. 167-228), is embedded in a manuscript from the cathedral, implying that this institution was a musical center equal in importance to Cambrai. A simple browse through the catalogue of special collections from the libraries and archives of the city, which was compiled before World War ii, uncovers numerous treasures once owned by these institutions.1 Many of these sources were destroyed in the war, and the few documents that survive today are housed in Tournai at the Archives de l’État, Bibliothèque de la Ville, and the Archives et Bibliothèque de la Cathédrale; and in Lille at the Archives du Nord. Only a scattering of manuscripts and printed books preserving the unique musical and devotional practices of Tournai cathedral exist in these collections (Pasture 1954).

Despite these limitations, it is possible to uncover a considerable amount of information pertaining to the music and liturgical practices of confraternities at Tournai cathedral. In particular, there is much evidence of these organizations’ patronage of polyphonic works, and the connection of their musico-liturgical practices to those of the cathedral at large and to other neighboring cities and dioceses. Given their musical legacy and proximity, it is not altogether surprising that certain practices might be shared between Tournai and Cambrai, but to date, few sources have come to light illustrating them. This study explores the vestiges of an improvised polyphonic tradition in the two dioceses, and shows how several confraternity manuscripts from Tournai and a fragment from Cambrai demonstrate this practice through their inclusion of three chants notated in cantus fractus: the Sanctus and Agnus Dei ‘Vineux,’ and Credo i. The

1. Faider (1950) lists many sources that were destroyed.

212 213 CANTUS PLANUS 2014 LONG — MENSURALLY NOTATED MASS ORDINARIES FROM CAMBRAI AND TOURNAI practice of notating chants in cantus fractus was prevalent in both musical centers, as chants represent different dissemination patterns between the dioceses of Cambrai the existing manuscripts attest. For example, the Médiathèque municipale in Cambrai and Tournai, and they show connections to other practices as far south as Paris, Dijon, has many sources containing chants in cantus fractus, among the most important and Avignon. being manuscripts F-CA MS 11 and MS 12 (Bouckaert 2007, pp. 92-94, 130). It is entirely possible (and probable) that these chants could have been performed without Several important themes intersect through an in-depth investigation of the contents embellishment, and simply as rhythmecized monophonic pieces. Their connection to of F-Lad MS 134 no. 12. First and foremost, this fragment has played an important role polyphonic settings, however, is undeniable. Several of the chants in cantus fractus in studies of Dufay’s biography and compositional practice, as he wrote a polyphonic are found elsewhere as tenors of polyphonic settings, with identical use of rhythm. setting of the Sanctus ‘Vineux.’ (Fallows 1976, pp. 44-50; Planchart 1993, pp. 341-368). In the case of Tournai, the existing liturgical sources reveal that polyphony – both To date, the fragment has only been discussed in this context, although it establishes improvised and written – was promoted to a large extent by confraternities at the a variety of intriguing connections to other subjects. For instance, its inclusion of cathedral, ultimately illuminating the many ways in which such communities were rhythmic chant notation is equally pertinent to work on hymn settings in cantus fractus active agents in the dissemination of new musical practices across dioceses. and improvised polyphony in manuscripts from Cambrai and Avignon (Sherr 1992, pp. 178-208; Ward 2004, pp. 331-343).5 Several scholars have done a great deal of research Credo i and the Sanctus and Agnus Dei ‘Vineux’ appear in several sources from northern on the rhythmic Credo i in Italy, and have found evidence that cantus fractus settings France and elsewhere, but both are found in different combinations in a series of three like the ones in F-Lad MS 134 no. 12, B-Tc A 12, and B-Tc A 27 are directly tied to local confraternity manuscripts used at Tournai cathedral. Bibliothèque de la cathedrale de improvised polyphonic traditions in the fourteenth century. These practices are thought Tournai manuscript A 12 and manuscript A 27 (hereafter: B-Tc A 12 and B-Tc A 27) to have first appeared in Avignon and made their way north to the Low Countries, and are two liturgical books used by the Confraternity of the Notaries at the cathedral south to Italy (Sherr 1992, pp. 178-208; Gozzi 2005, pp. 7-58; Gabrielli 2005, pp. 93-120; from the thirteenth through the sixteenth century.2 Both sources contain music for Gozzi 2006, pp. 245-302; Gozzi 2012a, pp. 137-154).6 Although F-Lad MS 134 no. 12 has the masses and offices observed by the confraternity at that time, as well as other never been linked to this custom, its use of mensural notation could place it seamlessly liturgical miscellanea. B-Tc A 27 includes the ‘Tournai Mass,’ discussed previously. within that tradition. When the Tournai manuscripts and the Cambrai fragment are Among the contents of B-Tc A 12 and B-Tc A 27 are identical versions of Credo i in examined together in respect to other sources from the region, a broader dissemination cantus fractus. Bibliothèque de la cathedrale de Tournai, manuscript A 58 (hereafter: pattern of rhythmecized plainchant emerges in relation to specific contexts. B-Tc A 58) contains masses, offices, and processions celebrated by the Confraternity of the Transfiguration at Tournai cathedral from the early fifteenth through the early The Sanctus ‘Vineux’ appears in a total of six sources, and it is the tenor of several seventeenth century. This manuscript has several polyphonic motet, mass ordinary, polyphonic settings. There are two manuscripts that contain polyphonic settings, and sequence settings, but most importantly for the present study, it includes the and four manuscripts that contain only the monophonic chant in cantus fractus. The Sanctus and Agnus Dei ‘Vineux’ in cantus fractus (Voisin 1853, pp. 139-144; Ceulemans polyphonic Sanctus and Agnus Dei ‘Vineux’ by Dufay, and the polyphonic Sanctus 2007, pp. 153-168; Pycke 2007, pp. 123-151; Ceulemans 2009, pp. 5-29). At first sight, ‘Vineux’ by Dufay’s presumed teacher, Richard Loqueville, are among the earliest and the musical traditions of the Confraternity of the Notaries and the Confraternity of most well known compositions to use the ‘Vineux’ melody as a tenor. Both settings the Transfiguration at Tournai cathedral appear to have no relationship to each other appear in Bologna, Museo Internazionale e Biblioteca della Musica di Bologna, MS Q when the three manuscripts described above are compared. These confraternity 15 (hereafter: I-Bc Q 15), which is the largest international compilation of polyphonic sources represent two overlapping traditions that are best illustrated through an works completed between 1420-1435 in the Veneto, Padua, and Vicenza. I-Bc Q 15 external source: a chant fragment currently held at the Archives du Nord in Lille, contains some pieces composed for the Council of Constance in 1415, and it is an MS 134 no.12 (hereafter: F-Lad MS 134 no. 12).3 This piece of parchment was used to important source for the works of Guillaume Dufay.7 Although I-Bc Q 15 is the most bind a will at the chapter of Sainte Croix in Cambrai, implying its origin there, and it prominent of all the sources for the polyphonic Sanctus and Agnus Dei ‘Vineux,’ an contains two chants in cantus fractus: Credo i and the Sanctus ‘Vineux.’4 Each of these anonymous three-voice setting of both chants appears in a fragment from Dijon, Bibliothèque Municipale, MS 2837(hereafter: F-Dm MS 2837) dated around 1420

2. Long (2011 and 2020) includes detailed inventories of the sources. 5. For a general study of hymns in cantus fractus and polyphonic settings, see Ward (1980). 3. I would like to thank Alejandro Planchart for sharing his unpublished inventory of this fragment with me. A brief inventory appears in Long (2020). 6. For transcriptions of mensural Credos, see Miazga (1976).

4. The Agnus Dei ‘Vineux’ is not included. 7. For a facsimile of the manuscript see: Bent (2008). For a full inventory and bibliography on the source, see Diamm.

214 215 CANTUS PLANUS 2014 LONG — MENSURALLY NOTATED MASS ORDINARIES FROM CAMBRAI AND TOURNAI

(Wright 1974, pp. 306-315).8 In both I-Bc Q 15 and F-Dm MS 2837, the tenor of the existing monophonic versions of the Agnus Dei each show different traditions, both Agnus Dei is a contrafactum of the Sanctus ‘Vineux’ chant. of which diverge significantly from each other and from the original Sanctus ‘Vineux’ melody. These two chants are found only in the confraternity manuscripts: B-Tc A 58 The monophonic Sanctus ‘Vineux’ circulated in cantus fractus in a total of four for the Confraternity of the Transfiguration at Tournai cathedral, and F-Pa MS 204 for sources, two of which may represent the original context for the chant. F-Lad MS 134 the Confraternity of Saint Sebastian at Notre Dame in Paris (table 2, p. 221). no. 12 is among the earlier manuscripts, dating at some point after 1393. It includes chants that formed part of the mass for the Unity of the Church, which was instituted A brief incipit of the Agnus Dei ‘Vineux’ in F-Pa MS 204 appears in example 1. When in that year by Pope Clement VII of Avignon (Amiet 1954, pp. 1-35). In addition to this version is compared to the one in B-Tc A 58, it is evident that there are similarities F-Lad MS 134 no. 12, the Sanctus also appears in Aosta, Seminario Maggiore, MS in the opening incipits. The first articulation of the Agnus Dei in B-Tc A 58 is almost 15 (hereafter: I-AO MS 15) which is an anthology of polyphonic music from the identical to the opening of the Sanctus, establishing it as a source for the initial melodic early fifteenth century that was compiled between 1430-1446 in Bologna, the Basel- material. Beyond that, the two renditions of the Agnus Dei ‘Vineux’ are unique, and Strasbourg area, and in Innsbruck. Like I-Bc Q 15, I-AO MS 15 also contains some have very little to do with the Sanctus melody. pieces performed in chapels connected to the Council of Constance (1414-1418) as Long_Arsenal 204 9 well as the Council of Basel (1431-1449). Although it is first and foremost a source Example 1. Opening of the Agnus Dei ‘Vineux’ (F-Pa MS 204). for polyphonic compositions, I-AO MS 15 contains only a monophonic version of the Sanctus ‘Vineux’ chant in cantus fractus. The remaining two manuscripts to include the Sanctus and Agnus Dei ‘Vineux’ are confraternity sources, both dating from the fifteenth and sixteenth centuries. In addition to B-Tc A 58 (discussed previously), the & ˙. ˙. ˙. ˙ ˙ œ ˙ œ œ A˙. -----gnusœ De˙. -˙i. . Pa - tris fi -- li us œ œ ˙. two chants appear in Paris, Bibliothèque de l’Arsenal, MS 204 (hereafter: F-Pa MS A - - - - - gnus De - i. Pa - tris fi - li - us______204), which is a late fifteenth- or early sixteenth-century liturgical book produced for the Confraternity of Saint Sebastian at Notre Dame in Paris (Long 2008, pp. 186- The Sanctus in B-Tc A 58 has an opening formula that proceeds in thirds to outline 2 10 235; Wright 2008 , pp. 134-139). a fifth above the starting tone, while the Agnus Dei in that source commences with a It is evident that the Sanctus ‘Vineux’ and its use of rhythmic notation are identical in the direct upward leap of a fifth. The version of the Agnus Dei in F-Pa MS 204 is identical monophonic and polyphonic (as a tenor) versions in all six sources, thereby showing to the one in B-Tc A 58 for the first four notes, but beyond that, the two chants are that it was transmitted in a very stable manner. Transcriptions of the Sanctus from completely different. The readings of both Agnus Dei chants may be considered local several different sources have been published, allowing for easy comparison (Besseler variants. One feature that clearly makes a difference is the trope, Patris filius eterni, 1962, pp. 9-10 and 1966, pp. 155-156; Wright 1974, pp. 306-315; Ceulemans 2009, pp. which appears in F-Pa MS 204, but not in B-Tc A 58. The Agnus Dei in F-Pa MS 204 11 15-16). The Agnus Dei, however, is different. The tenors of the two polyphonic settings proves considerably longer than the one in B-Tc A 58, since more notes are needed to of the Agnus Dei, one by Dufay in I-Bc Q 15 and the anonymous version in F-Dm MS accommodate the lengthy trope text.13 2837, are both contrafacta of the Sanctus melody (Wright 1974, pp. 306-315).12 The two When all of the versions of the Sanctus and Agnus Dei ‘Vineux’ are taken into account, 8. This is a series of fragments (ca. 1420) that were used to reinforce the binding of a late fifteenth-century incunabulum. it is clear that the Sanctus had a much more stable transmission than the Agnus Dei, It contains an anonymous three-voice setting of the polyphonic Sanctus and Agnus Dei ‘Vineux.’ Only the superius and final portion of the tenor of the Sanctus are preserved, and the Agnus Dei is poorly preserved. and that the latter was likely a tradition that came out of the polyphonic settings. The 9. Planchart (1993, pp. 359-360) has speculated that the polyphonic settings of the Sanctus and Agnus Dei ‘Vineux’ by Sanctus and the Agnus Dei ‘Vineux’ have strong connections to polyphonic traditions Dufay and Loqueville were composed for performance at the council. See Diamm for a full inventory and bibliography on the source. emanating from Cambrai due to their association with composers such as Dufay and

10. This manuscript contains masses in honor of Saint Sebastian, Saint Anthony, and Saint Geneviève. The only pieces Locqueville. Nevertheless, these traditions seem to have circulated in settings outside in cantus fractus are the Sanctus and Agnus Dei ‘Vineux.’ A full inventory and detailed discussion of the source appears in of the main sanctuaries of the cathedrals of Cambrai and Tournai. The ‘Vineux’ melody Long (2008). A preliminary discussion of the confraternity appears in Wright (2008). and its related polyphonic works appear only in manuscript compilations of polyphony 11. A complete transcription of the Sanctus in B-Tc A 58 appears in Ceulemans (2009). The legible parts of the polyphonic setting in F-Dm MS 2837 are transcribed in Wright (1974). The polyphonic settings by Dufay and Loqueville in I-Bc Q 15 are and confraternity sources. published in Besseler (1962 and 1966). The monophonic versions in I-AO MS 15, F-Pa MS 204, and F-Lad MS 134 no. 12 are identical to the tenors of the polyphonic settings in I-Bc Q 15 and the monophonic chant in B-Tc A 58.

12. See table 1 and table 2, p. 221, for a breakdown of the sources containing the Sanctus and Agnus Dei, and which ones include monophonic and polyphonic versions, and tropes. Wright was the first to note that the Agnus Dei is a contrafactum 13. This trope also appears in F-Dm MS 2837, but the tenor voice is too damaged to ascertain whether or not it matches of the Sanctus. F-Pa MS 204. The other voices of the Agnus Dei contain no match.

216 217 CANTUS PLANUS 2014 LONG — MENSURALLY NOTATED MASS ORDINARIES FROM CAMBRAI AND TOURNAI

The Credo i in cantus fractus that appears in F-Lad MS 134 no. 12 shows a different Overall, these are very small differences. A browse through Miazga’s index (1976) connection between Cambrai and Tournai in comparison to the Sanctus and Agnus and other published versions of the fourteenth-century Credo i from different Dei ‘Vineux.’ In addition to F-Lad MS 134 no. 12, the rhythmecized Credo i appears in geographical areas shows that various settings of the chant in Italy and elsewhere two manuscripts for the Confraternity of the Notaries at Tournai cathedral discussed are very similar to this one, making it difficult to ascertain how the rhythmic previously (B-Tc A 12 and B-Tc A 27), and in a missal and ritual currently held at Credo i appeared in Cambrai and Tournai (Gozzi 2005, pp. 25; Gabrielli 2005, pp. Tournai, Bibliothèque de la Ville, MS 13 (hereafter: B-Tv MS 13). This manuscript dates 107). Credo i is not written in rhythmic notation in any of the other sources from from the fourteenth century, and was likely used in a chapel at Tournai cathedral.14 this geographical area. Furthermore, its emergence primarily in confraternity manuscripts and those used in side chapels indicates that it was not part of the Credo i is one of the earliest settings of the Credo, and it appears in great numbers in performance practice in the main sanctuary at Tournai cathedral. Apart from F-Lad cantus fractus at the end of the fifteenth century and throughout the sixteenth century. MS 134 no. 12, the rhythmecized Credo i is not found in any other Cambrai sources This is linked primarily to the appearance of Credo iv in cantus fractus in numerous from the fourteenth or fifteenth centuries. This fragment was used to bind a will in sources of the time period, and a newly emerging performance practice in Italy that is the seventeenth century at the chapter of Sainte Croix in Cambrai, but aside from thought to have its roots in Avignon and other French territories (Gozzi 2006, pp. 245- this, there is no clear indication of its origins or its intended liturgical use, other than 246). While most of the mensurally notated settings of Credo i are found in manuscripts the fact that it appears with the mass for the Unity of the Church and the Sanctus and printed books from the late fifteenth century on, it was not unknown for it to ‘Vineux.’ As with B-Tc A 27 and B-Tc A 12, it is possible that it, too, originated within appear in fourteenth-century sources, such as those discussed here. a private devotional community.

Example 2 highlights certain portions of the beginning of Credo i in the Tournai and So far, this study has focused primarily on the transmission of the Sanctus and Agnus Cambrai sources. These versions are all nearly identical, with the exception of the Dei ‘Vineux,’ and Credo i, but the appearance of these mensurally notated chants cadences on ‘et terre’ and ‘invisibilium.’ In those instances, F-Lad MS 134 no. 12 and in sources from both Tournai and Cambrai says much about polyphonic practice B-Tv MS 13 are concordant, while the two confraternity manuscripts reveal a slightly at Tournai cathedral. Evidence of liturgical and musical similarities between the different practice. Variants like this are found throughout the chant melody.15 two dioceses is not at all surprising. The Scheldt (Escaut) river divides the city of Tournai, which was the seat of the diocese. Institutions on the eastern side of the river were subject to the diocese of Cambrai, and those to the west (including Example 2. Cadences in Credo i (F-Lad MS 134, B-Tv MS 13, B-Tc A 12, and B-Tc A 27) the cathedral) were subject to the diocese of Tournai. This proximity led to a considerable amount of crossover with regard to liturgical practices. For instance, F-Lad MS 134 the Confraternity of the Transfiguration, who owned and used B-Tc A 58 containing the Sanctus and Agnus dei ‘Vineux,’ was founded in the fifteenth century, before et ter - re______in - vi - si - bi - li - um 1445, at the parish church of Mont-Saint-Aubert in Tournai. This parish was, at the

B-Tv MS 13 time, under the jurisdiction of the diocese of Cambrai. Although the confraternity met at this institution, its membership, at a fixed number of eight, consisted of et ter - re______in - vi - si - bi - li - um priests who held benefices at Tournai cathedral. At some point between 1530 and

B-Tc A 12 and 1540, for reasons unknown, the confraternity moved its devotions from the parish B-Tc A 27 church to the cathedral (Pycke 2007, pp. 123-151). Thus, the Sanctus and Agnus Dei et ter - re in - vi - si - bi - li - um ‘Vineux’ came to Tournai cathedral by way of the diocese of Cambrai. Four of the six sources for the Sanctus and Agnus Dei ‘Vineux’ have some connection to Cambrai and composers active there, strengthening a case for its origin in this geographical location.16 On the other hand, the Credo i in cantus fractus appears primarily in 14. See Long (2020), which includes a full inventory of the source. This manuscript contains additions well into the sources from Tournai cathedral, with the only one from Cambrai being F-Lad MS fifteenth century, and includes the main feasts of the temporale, as well as some from the sanctorale. The manuscript at one time resided in the collection at the cathedral of Tournai, which is evident based on a pastedown on fol. 1: «Bibliothecae 134 no. 12. Such a setting could easily be tied to an improvised polyphonic practice ecclesiae Cathedralis Torn [sic]». There is no indication that the source was used by a confraternity, but due to its limited contents, it was likely used in a private devotional setting.

15. Although these two words appear together in the transcription, they do not appear in succession in the Credo. 16. These four manuscripts are F-Lad MS 134 no. 12; B-Tc A 58; I-Bc Q 15, and F-Dm MS 2837.

218 219 CANTUS PLANUS 2014 LONG — MENSURALLY NOTATED MASS ORDINARIES FROM CAMBRAI AND TOURNAI using the rhythmic plainsong in the tenor, as it was elsewhere in Italy and other Table 1. Appearances of the Sanctus ‘Vineux’ in manuscripts locations. Other chants in the Tournai manuscripts point toward such a tradition, for example, a rhythmically notated Kyrie from B-Tc A 12.17 trope text trope text Source Monophonic Polyphonic Qui ianuas mortis confregisti Qui vertice Thabor affuisti Both chants in cantus fractus imply an improvised polyphonic tradition at Tournai B-Tc A 58 cathedral, for which there is some precedent. As discussed previously, polyphony F-Lad MS 134 at Cambrai cathedral has been well studied, and the mensurally notated chants in the Lille fragment represent this tradition. In the case of Tournai, all of the sources I-Bc Q 15 explored in this paper were used in confraternal settings in the side chapels of the I-AO MS 15 cathedral rather than in the main sanctuary. To date, no foundations have come to F-Dm MS 2837 light mentioning polyphony from these organizations. Many references to singers and singing appear with regard to the main sanctuary of the cathedral, but rarely F-Pa MS 204 are there indications for polyphony. Perhaps the most famous foundation at the cathedral is from May 4, 1349, on which date the Bishop of Tournai instituted a mass for the Virgin Mary ‘cum nota’ to be sung in the main sanctuary. Jacques Table 2. Appearances of the Agnus Dei ‘Vineux’ in manuscripts Pycke has speculated that this was the occasion on which the polyphonic ‘Tournai Mass’ was originally sung, but there is no proof for that (Pycke 1988, pp. 51-57). One trope text Source Monophonic Polyphonic Patris filius eterni particular reference to polyphony in an extra-liturgical context at Tournai cathedral that is connected to the confraternity devotions discussed here is in relation to the B-Tc A 58 Annunciation drama performed in the main sanctuary on Ember Wednesday during I-Bc Q 15 Advent. At the end of the ceremony, the prosa Mittit ad virginem was sung with organ and discant (Young 1933, pp. 480-482).18 This chant circulated widely in F-Dm MS 2837 cantus fractus in France and the Low Countries, and its appearance in B-Tc A 27 in F-Pa MS 204 rhythmic notation indicates that a similar discant setting was also done within the confraternity.19

In the face of limited documents from Tournai, it is ultimately the manuscript sources containing music that speak to us the most. These clearly show the Confraternity of the Notaries and the Confraternity of the Transfiguration to be patrons of polyphonic works through their inclusion of polyphonic mass, motet, and sequence settings, and their use of mensurally notated plainchant.

17. A reproduction of the Kyrie appears in Long (2020).

18. Young (1933) gives a full transcription of the ceremony. It ends with the following instructions: «Item fiet Missa per omnia, ut in die Annunciationis Dominice cum sequentia sive prosa Mittit ad virginem, cum organis et discantu prout in triplicibus.»

19. Gozzi (2012b, pp. 259-264) gives a transcription of an alternatim setting in I-Bc Q 15, which shows even further connections between this source and the confraternity manuscripts discussed here.

220 221 CANTUS PLANUS 2014 LONG — MENSURALLY NOTATED MASS ORDINARIES FROM CAMBRAI AND TOURNAI

Bibliography Long Sarah Ann (2008), The Chanted Mass in Parisian Ecclesiastical and Civic Communities, 1480-1540: Local Liturgical Practices in Manuscripts and Early Printed Service Books, Ph. D. dissertation, Amiet Robert (1954), ‘La messe pour l’unité des chrétiens’, Revue des Sciences Religieuses 28, pp. 1-35. University of Illinois at Urbana-Champaign, pp. 186-235.

Bent Margaret (2008), Bologna Q 15: The Making and Remaking of a Musical Manuscript, Lucca: Lim – (2011), La musique et la liturgie de la confrérie des notaires à la cathédrale de Tournai à la fin du Moyen Age in Jacques Pycke ed., Archives et manuscrits précieux tournaisiens 4, Louvain-la Neuve: Besseler Heinrich ed. (1962), Guillaume Dufay. Fragmenta Missarum, Rome: American Institute of Université Catholique de Louvain (Tournai - Art et Histoire, Instruments de travail 15), pp. 45-66. Musicology (Cmm I/4). – (2020), Les dévotions des confréries à la cathédrale de Tournai du XIVeme au XVIeme siècle in Camilla – ed. (1966), Guillaume Dufay Compositiones Liturgicae Minores, Rome: American Institute of Cavicchi - Marie-Alexis Colin eds., Le Hainaut et la musique de la Renaissance: XVeme-XVIeme siècles, Musicology (Cmm I/5). Turnhout: Brepols.

Bouckaert Bruno ed. (2007), Cantus 21. Mémoires du chant. Le livre de musique d’Isidore de Séville à Miazga Tadeusz (1976), Die Melodien der einstimmigen Credo der römisch-katholischen lateinischen Kirche, Edmond de Coussemaker, Neerpelt: Alamire. Graz: Akademische Druck- und Verlagsanstalt.

Ceulemans Anne-Emmanuelle (2007), Le manuscrit musical Tournai B.C.T. A58 et les deux versions de la Molinier Auguste (1891), Catalogue Général des Manuscrits des Bibliothèques Publiques de France. Cambrai, séquence Thabor superficie in Anne Dupont - Jacques Pycke eds., Archives et manuscrits précieux vols. 66, Paris: Plon, vol. 17. tournaisiens 1, Louvain-la Neuve: Université Catholique de Louvain (Tournai - Art et Histoire, Instruments de travail 6), pp. 153-168. Pasture Abbé Alexandre (1954), ‘Inventaire des Archives du Chapitre Cathédral de Tournai’, Extrait de Archives, Bibliothèques et musées de Belgique 25, pp. 26-56, 185-219. – (2009), ‘Le manuscrit BCT A 58 de la cathédrale de Tournai et le plain-chant pour la confrérie de la Transfiguration’, Revue Belge de Musicologie 63, pp. 5-29. Planchart Alejandro (1993), ‘The Early Career of Guillaume Du Fay’, Journal of the American Musicological Society 46, pp. 341-368. Coussemaker Edmond de (1862), ‘Messe du XIIIe siècle’, Bulletins de la Société historique et littéraire de Tournai 8, pp. 100-110. Pycke Jacques (1988), Le culte de Notre Dame au XIVe siècle in Jean Dumoulin - Michel Huglo - Philippe Mercier - Jacques Pycke eds., La Messe de Tournai. Une messe polyphonique en l’honneur de Notre- Diamm. Digital Image Archive of Medieval Music Dame à la Cathédrale de Tournai au XVIe siècle. Étude et nouvelle transcription, Tournai-Louvain-la- http://www.diamm.ac.uk/ (accessed March 1, 2015). Neuve: Archives du Chapitre Cathédral, pp. 51-57.

Dumoulin Jean - Huglo Michel - Mercier Philippe - Pycke Jacques eds. (1988), La Messe de Tournai. Une – (2007), La confrérie de la Transfiguration au Mont-Saint-Aubert puis à la cathédrale de Tournai du 15e messe polyphonique en l’honneur de Notre-Dame à la Cathédrale de Tournai au xvie siècle. Étude et au 18e siècle in Jacques Picke ed., Archives et manuscrits précieux tournaisiens, 4, Louvain-la Neuve: nouvelle transcription, Tournai-Louvain-la-Neuve: Archives du Chapitre Cathédral. Université Catholique de Louvain (Tournai, Art et Histoire 25; Instruments de travail 6), pp. 123-151.

Faider Paul (1950), Catalogue des manuscrits conservés à Tournai (Bibliothèques de la Ville et du Séminaire), Sherr Richard (1992), The Performance of Chant in the Renaissance and its Interactions with Polyphony Gembloux: Duculot. in Thomas Forrest Kelly ed., Plainsong in the Age of Polyphony, Cambridge: Cambridge University Press, pp. 178-208. Fallows David (1976), ‘Dufay and Nouvion-le-Vineux: Some Details and a Thought’, Acta Musicologica 48, pp. 44-50. Voisin Charles-Joseph (1853), ‘Notice sur la confrérie de la Transfiguration’, Bulletins de la Société historique et littéraire de Tournai 3, pp. 139-144. Gabrielli Giulia (2005), Il manoscritto 327 della Fondazione Biblioteca San Bernardino di Trento in Gozzi and Luisi 2005, pp. 93-120. Ward Tom R. (1980), The Polyphonic Office Hymn 1400-1520: A Descriptive Catalogue, Neuhausen-Stuttgart: American Institute of Musicology-Hänssler. Gozzi Marco (2005), Il canto fratto: prima classificazione dei fenomeni e primi esiti del progetto RAPHAEL in Gozzi and Luisi 2005, pp. 7-58. – (2004), Polyphonic Settings of Mensurally Notated Hymn Melodies in Andreas Haug - Christoph März - Lorenz Welker eds., Der lateinische Hymnus im Mittelalter, Kassel: Bärenreiter, pp. 331-343. – (2006), ‘Alle origini del canto fratto: il Credo Cardinalis’, Musica e storia 14, pp. 245-302. Wright Craig (1974), ‘A Fragmentary Manuscript of Early 15th-Century Music at Dijon’, Journal of the – (2012a), I prototipi del canto fratto: Credo regis e Credo cardinalis in Marco Gozzi ed., Cantus fractus American Musicological Society 27, pp. 306-315. italiano: un’antologia, Hildesheim: Olms, pp. 137-154. – (20082), Music and Ceremony at Notre-Dame of Paris, 500-1550, Cambridge: Cambridge University – (2012b), La sequenza Mittit ad Virginem recostruita da Tr 92 in Marco Gozzi ed., Cantus fractus Press. italiano: un’antologia, Hildesheim: Olms, pp. 259-264. Young Karl (1933), The Drama of the Medieval Church, Oxford: Clarendon Press. Gozzi Marco - Luisi Francesco (2005) eds., Il canto fratto: l’altro gregoriano. Atti del convegno internazionale di studi, Parma-Arezzo, 3-6 dicembre 2003, Roma: Torre d’Orfeo.

Guletsky Irene (2009), ‘The Four 14th-Century Anonymous Masses: Their Form; the Restoration of Incomplete Cycles; and the Identification of Some Authors’, Acta Musicologica 81, pp. 167-228.

222 223 Hana Vlhová-Wörner

Agnus pairing and disappearing: a contribution to the late chant tradition in Bohemia

I The prolonged continuation of the chant tradition has recently attracted a new interest in medieval scholarship, with publications addressing a number of important topics, such as liturgical modifications, rhythmical interpretation of ‘plainchant’ melodies, or a different compositional approach during the time that has been primarily devoted to polyphonic music (Karp 2005; Gozzi 2012). Late chant repertory has been particularly associated with the fifteenth- and sixteenth-century music tradition in Bohemia, where the religiously motivated Hussite wars (1419-1434) ended in the constitution of the first religious fraction widely independent from Rome. Bohemia had already transformed in the second half of the fifteenth century to a unique music-liturgical landscape, marked by various interests in the repertory (chant, archaic polyphony, fifteenth-century polyphony, strophic spiritual song) and by the language used in the liturgy (Latin, vernacular, Latin and vernacular). Despite the distinctive fragmented developments, monophonic chant remained present in Bohemia in the liturgy of both main religious fractions – the church sub una (which remained subordinated to Rome) and the genuine Bohemian church of Utraquists (practicing the communion of both Body and Blood – sub utraque specie), not only as a music idiom truthfully preserved and occasionally adjusted to new aesthetic criteria and liturgical environment, but also as a subject of prudent modifications and numerous additions.1 Writings on Utraquist chant in Bohemia, the earliest of them dating from the 1920s,2 have already recognized a number of characteristic repertory changes introduced to the chant (and partly also polyphonic repertory) by the end of the fifteenth century, such as the performance of abbreviated Creed chants, resulting from the distance of the Utraquist church from its final clause (et unam sanctam), or the introduction of short, three-fold tropes to the Kyrie eleison (Vlhová-Wörner 2008).3 Of all the characteristic features, the sudden disappearance of Agnus from collections of the mass ordinary chants around 1500 has always been to the most puzzling and has become in recent years the subject of a renewed discussion, involving music historians, liturgists and art historians (Koláček 2008; Žůrek 2009; Horyna 2011). The circumstances surrounding this peculiar development are far from clear. Many representative Utraquist chant books, such as

1. On the history of the Utraquists church in Bohemia, see David (2003).

2. Dobroslav Orel (1922) includes not only transcriptions of selected pieces and compositions from this representative Utraquist chant book from Králové, but also the first comprehensive observations on the Utraquist liturgical music.

3. Edition of abbreviated Kyrie eleison chants is included in Vlhová-Wörner (2006).

225 CANTUS PLANUS 2014 VLHOVÁ-WÖRNER — AGNUS PAIRING AND DISAPPEARING: THE LATE CHANT TRADITION IN BOHEMIA the Franus Kancionál from 1505 (CZ-HK II A 6), its closely related Graduál-Kancionál simply borrowed from the collection of preceding Sanctus chants (Orel 1922, p. 36). of Martin Baccalarius of Vyskytná from 1512 (CZ-Pn XIII A 2), or the beautifully The discussion so far has focused primarily on the evaluation of Bohemian sources decorated late-fifteenth century graduál-kancionál from Kutná Hora (A-Wn 15501) from the last decades of the fifteenthand the sixteenth centuries. In this contribution, I do, indeed, finish their ordinary chants collections with the Sanctus section. All three would like to follow another path and describe the whole phenomenon as a final stage manuscripts have traditionally been included in the main focus in the study on the late of a much longer process that started well before the outbreak of the Hussite wars and Utraquist repertory, be it monophonic or polyphonic.4 This explains why generations establishment of the Utraquist church, namely in the late fourteenth century. I will of scholars accepted the ‘missing Agnus Dei’ phenomenon as a general characteristic of demonstrate that the process included not only new arrangements within the mass all (Latin) Utraquist chant books. Only recent systematic research on the fifteenth- and ordinary collections, but also new compositions, musical elaborations and, perhaps, sixteenth-century Latin and vernacular chant books, which for the first time also took also adoption of a new genre in order to enrich and later to replace the traditional into consideration smaller (local) manuscript libraries in Bohemia, created a serious liturgical form. crack on this formerly smooth picture, by identifying dozens of late chant books that contrary to previous knowledge include at least a modest collection of Agnus melodies II (Graham 2006).5 Late-medieval Bohemian graduals provide enough evidence to support the ‘economic’ explanation already proposed by Orel. Changes in the organization of the mass Such a contradictory finding certainly has liturgical rather than musical implications, ordinary chants in Bohemian manuscripts towards the end of the fourteenth century suggesting that there was no revised or even ‘reformed’ liturgical practice in Bohemia nourish the idea that Sanctus and Agnus chants became in addition to the ‘ordinary’ that would have been respected by all Utraquist churches. In attempting to explain (unchangeable) character of their core texts closely connected also by their melodies. the differing profile of Utraquist chant books, scholars involved in the recent While ordinary chants in older collections were typically arranged in groups (Kyrie, discussion have looked for specific liturgical practices and influences external to the Gloria, Sanctus, and Agnus), with occasional insertions of Gloria incipits that should be Utraquist context. Currently the most popular explanation is that the frequent Agnus attached to one particular Kyrie melody,6 manuscripts from ca. 1400 document another disappearance from music manuscripts must be put into a broader sixteenth-century arrangement, namely a formation of larger repertory sections with pairs (Kyrie and development, which was generally marked by the abbreviation of the second half of Gloria; Sanctus and Agnus) that are musically either thoroughly identical or largely the mass liturgy. Not only Agnus chants, but also other liturgical chants – such as similar. Comparable arrangements of some mass ordinary chants can be observed Sanctus, offertory and communion – gradually disappear, possibly making space for elsewhere since the thirteenth century, yet never in such degree as in fifteenth- other repertory genera: spiritual songs and polyphonic pieces (Horyna 2011, p. 19). and sixteenth-century Bohemian manuscripts.7 The consistency, with which some The simplification of music repertory could be understood as the fruit of the influence Bohemian compilers undertook their plan to build pure ‘semi–cyclic’ series (series of the just established Lutheran church (Koláček 2008, p. 10). But, on the other hand, with Kyrie-Gloria and Sanctus-Agnus pairs) through elaborations, adjustments, or additional liturgical sources indicate that Sanctus and Agnus should both remain part even new compositions, is remarkable. Late collections of ordinary chants included in of the mass liturgy, but only in the form of the silent recitation that was reserved for manuscripts often leave only a little space for musically divergent pairs. The gradual the priest (Horyna 2011, pp. 18-20). Another possibility is that the ‘reduced’ repertory from the central-Bohemian city Kouřim from 1470 (CZ-Pu XIV A 1), representing the reflected the ceremony of the morning mass, for which the Literary Brotherhoods – the sub una liturgical practice, can be introduced here as a typical example. Its collection owners of many Utraquist chant books – were responsible (Horyna 2011, pp. 16-20). of ordinary chants, inserted as usual at the beginning of the manuscript (fols. 2-24), All these recent speculations cast doubt on the once preferred explanation offered by contains a fairly large series of twenty Sanctus-Agnus pairs, of which only four cannot Dobroslav Orel, who understood Utraquist liturgy as a static protocol closely related be safely labeled as ‘fully’ or at least ‘largely’ identical; but even these four pairs in to traditional Roman liturgy and looked for a more practical reason to explain the question are connected at least by the same mode (D or F). Several approaches can missing Agnus repertory. According to him, despite their omission in the manuscript be recognized from the musical unification of Sanctus and Agnus chants. Evidently, the Agnus chants were performed during the liturgy, whereby their melodies were the favored technique was to borrow an established Sanctus melody for a ‘new’ Agnus

4. In addition to the Orel’s monograph, the Franus kancionál was included as a pilot manuscript with the polyphonic music 6. This organization is followed in the so-called Saint Vitus’s troper from 1235 (CZ-Pak Cim 4) or in the representative into the catalogue Rism B IV3 (pp. 141–163) and was also mentioned as one of the representative ‘Kantional’-manuscripts collection with mass ordinary chants from the Saint Vitus’s cathedral, the so-called Ordinarium of the Archbishop Arnestus in the article Kantional in Mgg1 from 1958 (vol. 7, col. 611). A detailed repertory list with further concordances is available of Pardubice from 1360s (CZ-Pak P IX). in Černý (1966). 7. Hiley (1986) includes a handful of comparable instances, as, for example, in the Parisian notated missal from the second 5. A more representative selection is included in the database Limup (http://clavmon.cz/limup) developed by J. Žůrek half of the thirteenth century (London, British Library, GB-Lbl Add. 38723) or in the thirteenth-century gradual from Rouen since 2002. (Paris, Bibliothèque Nationale, F-Pn fol. l. 904); See also Kiss (2009, pp. 25-26).

226 227 CANTUS PLANUS 2014 VLHOVÁ-WÖRNER — AGNUS PAIRING AND DISAPPEARING: THE LATE CHANT TRADITION IN BOHEMIA chant: several pairs in the series were created through this simple adaptation.8 Also Table 1. Tropes to the Sanctus (Sa) and Agnus Dei (Ag) in gradual CZ-Pnm XIII E 8, Bohemia (ca. 1400) possible, though less often applied, was the opposite approach, when the Sanctus melody was borrowed from an established Agnus chant. This can be observed in one No. Folio Trope Incipit Rubric pair, where the Sanctus (Thannabaur 1962, 110),9 documented since the fourteenth 1 119v-120 Sa Deus pater, iudex In summis festivitatibus century almost exclusively in Bohemian manuscripts, quite clearly quotes the widely distributed early Agnus melody (Schildbach 1967, 114; Vat. IX). The genuine late- 2 120r-v Ag Danielis prophecia medieval technique, however, observed in a number of additions to the older repertory 3 120v Sa Deus, rerum principium Item summum of the mass ordinary chants, was to compose a new, musically ‘identical’ or ‘largely identical’ couple: this, for example, was the case with the pair Sanctus (Thannabaur 4 120v-121 Ag Reparator vite 1962, 39) and Agnus (Schildbach 1967, 56) that are documented in the Bohemian 5 121r-v Sa Genitor summi filii manuscripts exclusively. 6 121v Ag Qui de celis descendisti Sequens Sanctus canitur The idea of melodic unity between Agnus and Sanctus, recognizable in a number of 7 122 Sa O quam dulciter voces Bohemian manuscripts already in the late fourteenth century, logically provided the in festo Omnium sanctorum possibility to apply one melodic formula to both Sanctus and Agnus chants. Since they 8 122r-v Ag Rex eterne glorie both follow a three-fold form and are built on the same principle of repetition, it was 9 122v-123v Sa Divinum mysterium De Corpore Christi certainly not difficult for singers to create the required sister chant. Orel was able to find support for his hypothesis in the gradual from Slavětín from the beginning of 10 123v Ag Fons indeficiens pietatis the sixteenth century, where a scribe added the text of the Agnus above the notated 11 123v-124 Sa Constat in altari Item aliud de Corpore Christi […] Sanctus, certainly as a simple hint for the performers (Orel 1922, p. 36). 12 124 Ag (without tropes) Hoc Sanctus canitur It remains the exclusive domain of late Bohemian chant repertory that the process of 13 124 Sa Flos candens oritur in Nativitate beate virginis pairing affected not only melodies themselves, but also tropes to the Sanctus and Agnus Dei. A recent critical edition of the mass ordinary tropes from Bohemian manuscripts 14 124r-v Ag O flos regalis identified several instances of such elaborations, which aimed to unify musically the In summis festis sive 15 124v-125 Sa Genitor ingenitus Agnus trope with the trope to the Sanctus prescribed for the same liturgical occasion in maioribus De martiribus in festis maioribus, in the manuscript. Again, there is no indication that these elaborations were connected 16 125 Sa Deus, rerum principium (bis) quando celebrantur with a repertory ‘revision’ representing a monastic/secular community in the pre- Hussite period, or sub una/Utraquist church in the fifteenth century and beyond. As 17 129r-v Sa Angeli et archangeli Item hoc canitur, quando placet they can be found across a wide spectrum of manuscripts and in collections of different Istud Sanctus canitur 18 129v-130 Sa Salve, mater pia profiles, they must be interpreted differently, namely as an expression of the idea that in festivitatibus beate virginis both chants can or even perhaps should appear as members of one larger complex, a concept that is otherwise associated primarily with fifteenth-century polyphonic tradition. Among several manuscripts from the pre-Hussite period, a Bohemian For its most part, the mass ordinary chants in the collection are systematically arranged gradual of unknown origins (possibly Augustinian?) from ca. 1400 displays the highest in pairs, with a section containing a series of Kyrie-Gloria pairs (not shown in the degree of such reorganization, connected with several elaborations of Agnus tropes table) and another section with Agnus-Sanctus pairs.10 Chants with tropes are here in melodies (table 1). accordance with the common practice prescribed for the most important feasts of the liturgical year (In summis festivitatibus, De Corpore Christi, etc.), placed mostly in the first half of the series. The set of ordinary chants still continues in the manuscript after Sanctus Salve, mater pia (table 1, no. 18) with non-troped Sanctus and Agnus chants, 8. This approach has been recognized already by Martin Schildbach (1967, pp. 51-52): one third of Agnus melodies included in his catalog is either identical or very similar to Sanctus melodies.

9. Melodies of Sanctus and Agnus are identified here by their numbers given in the following catalogues: Thannabaur 10. Confer similar arrangement in the gradual from the All Saints church in Prague Castle from ca. 1400 (D-Mbs 23286, 1962 and Schildbach 1967. fols. 2–30v).

228 229 CANTUS PLANUS 2014 VLHOVÁ-WÖRNER — AGNUS PAIRING AND DISAPPEARING: THE LATE CHANT TRADITION IN BOHEMIA the latter in a simple polyphonic setting.11 With a single exception of tropes for the feast prescribed for the same important feast of the church year that have otherwise nothing of All Saints (Sanctus O quam dulciter voces; Agnus Rex eternum glorie, table 1, nos. in common: while the trope Divinum mysterium narrates the Consecration of Bread 7 and 8), all trope pairs are linked either by the same mode, the same music incipits, and Wine and was typically intended for the Corpus Christi feast and the Maundy or they share the same or widely similar melody, even if the correspondence required Thursday in Bohemian manuscripts, the text of the trope Fons indeficiens, as many some modifications or a completely new musical setting. representatives of this genre, is largely neutral.

The latter is clearly the case with the trope Agnus Fons indeficiens pietatis(Vlhová- Another pair of tropes, prescribed in the collection for the feast of the Nativity of the Wörner 2013, AgTr 5). Within the trope’s tradition that started ca. in 1100 in the Holy Virgin – Sanctus Flos candens oritur and Agnus O flos regalis virgule (table 1, west and continued since the thirteenth century also in the east of the Rhine, it nos. 13 and 14) – displays a different approach. Flos candens oritur belonged in the was transmitted in two versions, both of them, however, share the same mode (F fourteenth century to the established Marian trope repertory of the central-European or, in most cases, its transposition in G) and the overall melodic contour: the open territory, as it can be found in a number of the fourteenth-century south-German and cadence on a/b at the end of each strophe, which prepares a continuous transition Bohemian manuscripts.12 The trope O flos regalis virgule, on the contrary, survived in to the final supplication miserere nobis is characteristic. A distinctive feature of the two manuscripts only. Apart from the Bohemian gradual CZ-Pn XIII E 8, it is included ‘eastern’ tradition is ornamental embellishment on the opening phrase (Fons) and also in the Codex Engelberg from the late fourteenth century, where it immediately the supplementary repetition of the closing supplication miserere [nobis] that can follows, similarly to the Bohemian gradual, the trope Sanctus Flos candens oritur be understood as the extended final line of each stanza. The ‘western’ version, on (fol. 108v–109).13 There are several reasons to believe that these two chants were also the contrary, is characterized by a more narrative melody proceeding in syllabic considered in the Codex Engelberg as a pair: they are connected by the characteristic movement with occasional insertions of short embellishments; this is also the version, word flos in their opening text lines; both are composed in the F mode and, most with additional melismas, that reached Bohemia at latest in the mid-thirteenth century. importantly, their melodies display a similar use of ornamentation, introducing a long This version is included, together with other western ordinary tropes, in the collection melisma at the beginning and end of each stanza (see example 1). of the mass ordinary tropes from the end of the twelfth century, in the so-called Saint Vitus’s troper purchased in 1234 for the Prague Cathedral chapter, and was copied Example 1. Engelberg, Stiftsbibliothek, Codex 314, f. 108v again in the 1360s into the representative manuscript of the mass ordinary chants procured by Prague’s Archbishop Arnestus for the Saint Vitus’s canons. The gradual CZ-Pn XIII E 8 constitutes the only witness that this trope found its way from the Saint Vitus’s Cathedral to the neighboring churches. The text form avoids the supplementary repetition miserere nobis, thus matching the trope’s western design: this detail alone excludes the possibility that the trope was received from the neighboring south German territory that otherwise would have supplied Prague’s diocesan churches with new repertory. The trope melody in the gradual CZ-Pn XIII E 8 is, however, clearly different from both east and west traditions, presenting an isolated departure from otherwise relatively constant transmission. It was not newly composed, but borrowed together with its ordinary chant melody from the Sanctus Divinum mysterium (Vlhová-Wörner 2010, SaTr 15), which is here coupled with the Agnus Fons indeficiens prescribed for the Neither of these elements is fully sufficient to create a link between both chants and it Corpus Christi feast. Without knowledge of the parallel development in the repertory is less important that the opening ornamental melisma of O flos regalis virgule is here of the (non-troped) mass ordinary chants described above, it would be very difficult clearly shorter in comparison with the large melodic arc covering the full ambitus to explain the decision to exchange a well-established melody for a new one. In the of the authentic and plagal F mode at the beginning of the trope Flos candens oritur. broader context, however, the reason behind this sudden departure from the tradition Yet, the relationship between both chants is expressed much more strongly in the is quite clear: the intention was to unify both tropes – Sanctus Divinum mysterium Bohemian gradual, where both tropes appear literally as musical twins: not only the and Agnus Fons indeficiens – by their melodies, in order to create a pair of chants

12. Vlhová-Wörner 2010, SaTr 21.

11. Rism B IV3, pp. 231–235 (ms. CZ–Pn XIII E 8), nos. 1 and 2. 13. Vlhová-Wörner 2013, AgTr 9.

230 231 CANTUS PLANUS 2014 VLHOVÁ-WÖRNER — AGNUS PAIRING AND DISAPPEARING: THE LATE CHANT TRADITION IN BOHEMIA opening melisma, but the melodies of their stanzas are fully identical. The unification of liturgical poetry, a subject of many transformations and compositional experiments. of both chants went as far as to modify also the selection of mass ordinary melody In Bohemian manuscripts tropes to the Agnus are generally smaller in number, in the Bohemian gradual. The trope Flos candens oritur is here uniquely attached to when compared to tropes to the Sanctus; this can also be observed in the gradual the melody no. 106 of Thannabaur catalogue (1962); the same melody can be found CZ‑Pn XIII E 8 mentioned above that includes, in addition to six Sanctus-Agnus pairs, attached to the Sanctus Divinum mysterium) and its musical equivalent, the melody four more tropes to the Sanctus at the end of the series (table 1, nos. 15-18) and the no. 124 of Schildbach catalogue (1967) appears with the trope O flos regalis virgule. trope Sanctus Constat in altari linked to an untroped Agnus (table 1, nos. 11-12). As can be seen in table 2, the most representative fourteenth-century Bohemian manuscript Only in this context, it is possible to explain why a small melodic modification also with ordinary chants, the so-called Ordinarium Arnesti from 1360s, displays a similar affected the widely known trope Danielis prophecia, documented since the thirteenth ratio: the manuscript includes no less than eighteen tropes to the Sanctus – one of the century in numerous manuscripts in the south German tradition and later in central highest number in the European context –, but only eleven tropes to the Agnus. Europe.14 Together with the prime Bohemian trope to Sanctus Deus pater, iudex iusticie, it opens the series of the mass ordinary tropes in the gradual CZ-Pn XIII E 8, where it Table 2. Tropes to the Sanctus and the Agnus Dei in Bohemian collection: a statistics is prescribed for the main feasts of the Church year (table 1, nos. 1 and 2). While the Tropes Tropes melody of the trope Danielis prophecia starts with a fifth-up jump (d–a) in German Manuscript to Sanctus to Agnus Dei manuscripts, it is filled with an additional third, f, in the Bohemian gradual. Even if the modification is rather small, as in this case, the resulting melodic gesture is very Ordinarium Arnesti de Pardubice, CZ-Pak P IX (ca. 1360s) 18 11 similar and it can be surely interpreted as an attempt to adjust the beginning of the Prague University (?) Collection, CZ-Pu V H 11, (ca. 1380s) 8 0 trope to the preceding Sanctus Deus pater, iudex iusticie in order to create a musical Gradual from Bohemia (OSA?), CZ-Pn XIII E 8 (ca. 1400) 11 7 affinity between both chants. All these examples demonstrate a growing notion of close melodic similarity between Sanctus and Agnus and strengthen Orel’s proposed ad hoc Gradual from the All Saints Church, D-Mbs 23286 (ca. 1400) 7 1 performance of unwritten Agnus Dei chants. They brought, however, an important Collection from Vyšší Brod [Hohenfurth], CZ-VB 42 (1410) 9 1 correction as for the definition of the period, in which this important rearrangement within the mass ordinary repertory started: not in the last decades of the fifteenth Vyšehrad Collection, CZ-Pst KVš 376 (ca. 1460) 7 1 (+1) century, but already before the outbreak of the Hussite wars. Advanced organization Gradual from Kouřim, CZ-Pu XIV A 1 (ca. 1470) 4 1 of fifteenth-century mass ordinary chant collections may present in this context a progressive stage of a much longer process, with a group of manuscripts containing faithfully the records of all matching pairs and another group of manuscripts, where Yet, some manuscripts from the same period typically include a reasonably large scribers opted to leave the Agnus repertory completely out. Because Utraquist churches series of tropes to the Sanctus, still with occasional new additions or elaborations, but, did not observe a unified repertory in the late fifteenth and sixteenth centuries, there strangely, a very limited number of tropes to the Agnus. First of all, two representative still remained the possibility to preserve in the chant books a reasonable repertory of sources of the Prague repertory from the beginning of the fifteenth century, namely the Sanctus and Agnus, melodically similar or different. gradual from the All Saints church in Prague Castle (D-Mbs 23286) and the so-called Vyšebrod collection from 1410 (CZ-VB 42), include a series of tropes to the Sanctus III (seven and nine, respectively), but only one single trope to the Agnus. Moreover, The most recent study of the mass ordinary chants and its tropes in Bohemia has the trope in question, Agnus Miserere deus vere, appears only in the late fourteenth revealed, however, that there is another important movement within the late medieval century central European region as a rare representative of an Agnus prosula having mass ordinary repertory so far not recognized: the sudden drastic reduction followed stylistically only little in common with the fourteenth-century repertory of the Agnus by a later disappearance of tropes to the Agnus observed already at the turn of the tropes. Moreover, the university collection CZ-Pu V H 11, famous for its focus on the fourteenth and fifteenth centuries. This development is quite surprising, not in the late medieval liturgical poetry, contains eights tropes to the Sanctus but omits tropes to least because the sister tropes to the Sanctus remained an important part of the the Agnus completely (Plocek 1973, pp. 103-117). Several manuscripts from the second liturgical repertory until the late Middle Ages and, perhaps more than any other genre half of the fifteenth century have a similar profile, whereby all of them represent either the liturgy of the sub una church or come from monastic communities. The 14. Vlhová-Wörner 2013, AgTr 3; Hospental 2010. Kouřim gradual from 1470, which was introduced at the beginning of this paper as

232 233 CANTUS PLANUS 2014 VLHOVÁ-WÖRNER — AGNUS PAIRING AND DISAPPEARING: THE LATE CHANT TRADITION IN BOHEMIA a manuscript with an elaborated series of musical Sanctus-Agnus pairs, displays the As already explained elsewhere, the smooth functional exchange between Sanctus same tendency, with four tropes to the Sanctus and, again, only one single prosula to and Agnus tropes, exceptionally applied for the enrichment of the Agnus tropes the Agnus (Miserere deus vere). repertory (that is, absorbing chants from the Sanctus tropes family), was possible only after the standard three-fold strophic form of Sanctus and Agnus tropes was This does not mean, however, that tropes to the Agnus transmitted in the fourteenth established in the fourteenth century Central European territory and the tropes century in Bohemia disappeared without any trace from the repertory. Some of them became independent strophic chants, closely related to the late medieval Latin were preserved in the ordinary tropes collections, transformed – usually without any cantio.17As a consequence, the bond between the trope and its ordinary chant was significant changes in their texts or melodies – into the tropes to theSanctus chants often free in the late tradition. This also demonstrates the rubric before the trope (Table 3).15 Rector celi inmortalis in the so-called Vyšehrad collection from ca. 1460, which informs us that the piece can be used both as the Sanctus and Agnus trope (CZ-Pst Table 3. Agnus Dei tropes (AgTr) with the ‘dual function’ in late medieval Bohemian tradition KVš 376, fol. 125v: Trophus super Sanctus and Agnus). Only two opening Sanctus invocations are attached to the inscription and it is left to the performers to apply the

Liturgical function relevant melody of the Agnus chant: a situation strikingly similar to Orel’s scenario for explaining the missing Agnus repertory in some late Bohemian manuscripts! Incipit Outside Bohemia In Bohemia

Ave, dulce tu frumentum (Vlhová-Wörner 2010, SaTr 3) none AgTr " SaTr IV Taking all observations together, we get a picture of a surprisingly dynamic Constat in altari (Vlhová-Wörner 2010, SaTr 7) AgTr AgTr " SaTr movement within the mass ordinary repertory in the Bohemian tradition that started Eructavit cor meum (Vlhová-Wörner 2010, SaTr 18) AgTr " SaTr latest by the end of the fourteenth century (table 4). The growing importance of the musical affinity between Sanctus and Agnus Dei, observed foremost in manuscripts O Maria, tu cum agno (Vlhová-Wörner 2010, SaTr 30) AgTr " SaTr coming from the church sub una, can be most probably attributed to the influence Qui de celis descendisti (Vlhová-Wörner 2013, AgTr 11) AgTr AgTr " SaTr of the parallel development within the polyphonic repertory in the second half of 18 Rector celi inmortalis (Vlhová-Wörner 2010, SaTr 41) SaTr AgTr and SaTr the fifteenth century, which is marked by the establishment of the cyclic masses. Yet, how much this idea was planted in the monophonic repertory in the preceding period is still to be answered. It is evident that the repertory of the Sanctus tropes attracted new additions borrowed from its sister chants transmitted in Bohemia or adjacent regions already in the late Table 4. Changes within the repertory of Sanctus and Agnus Dei in late medieval Bohemia fourteenth century: the gradual from the All Saint church in Prague Castle (D-Mbs 23286), for example, includes in its repertory the trope Eructavit cor meum (not known Church sub una Utraquist church from other Bohemian manuscripts), documented otherwise as a trope to the Agnus 1500 Sanctus / Agnus melodic pairs No Agnus chants (occasionally) 1470 in manuscripts from Spain, north France and south Germany, from where it most Small Agnus tropes repertory probably was received.16 It was perhaps the character of its text consisting of quotations 1450 Period of the reestablishment 1420 Hussite wars (1419-1434) borrowed from other liturgical chants, which allowed this haphazard exchange within the liturgical repertory. A different case presents, however, the transfer of the trope O New organization Agnus tropes transformed to Sanctus tropes Maria, tu cum agno from the family of Agnus tropes to the repertory of Sanctus. The 1400 of mass ordinary Decreasing number of Agnus tropes collections Agnus tropes recomposed / unified with Sanctus tropes text of the opening stanza clearly correlates with the text of its original mass ordinary chant (Cum agno and Agnus Dei), but this link is broken after the same text was adapted for the Sanctus addition.

15. Vlhová-Wörner 2013, pp. 29–30. 17. Vlhová-Wörner 2013, p. 29. 16. Vlhová-Wörner 2010, SaTr 18. 18. See, for example, Fischer (1964).

234 235 CANTUS PLANUS 2014 VLHOVÁ-WÖRNER — AGNUS PAIRING AND DISAPPEARING: THE LATE CHANT TRADITION IN BOHEMIA

The well-known and many times discussed disappearance of Agnus repertory from List of Manuscripts some representative Utraquist chant books was without any doubt possible only because of the autonomous statute of the Utraquist church and its liturgy, which A-Wn 15501 remained independent from the centralised Roman ritual. Absence of such an Vienna, Österreichische Nationalbibliothek, ms. 15501; Kutná Hora, end of the fifteenth century. important repertory group in a liturgical book with the official ‘Roman’ liturgy is CZ HKm II A 6 hard to imagine (and was, so far, never observed outside Bohemia). It is also evident Hradec Králové, Museum of Eastern Bohemia, ms. II A 6; Hradec Králové, 1505 that this particular repertory characteristic has only little to do with radical changes CZ-Pak Cim 4 within the mass liturgy later in the sixteenth century, which could be influenced, as Prague, Archive of Prague Castle, Library of the Metropolitan Chapter at St Vitus, Rism ms. Cim 4; origin already suggested, by the newly established Lutheran church. As the table clearly unknown (from 1234 in the Library of the St Vitus’s Chapter), 1180–1200 demonstrates, the missing Agnus in some Utraquist books is only one of several CZ-Pak IX indications that the focus on Agnus repertory – be it its tropes or its melodies that were Prague, Archive of Prague Castle, Library of the Metropolitan Chapter at Saint Vitus, ms. P IX; Prague, Saint Vitus’s Cathedral, 1360–1370 now often derived from the Sanctus chants – generally decreased in late medieval Bohemia. Striking is that the change in the repertory developed parallel to the growing CZ-Pn XII A 1 importance of the second half of the mass and the act of the communion, which became Prague, National Museum Library, ms. XII A 1; Prague, Saint Vitus’s Cathedral, end of fourteenth century + a. 1473 at the beginning of the fifteenth century one of the most crucial topics associated with the pre-Reformation movement. Everything points out that it was a practical reason CZ-Pn XIII A 2 Prague, National Museum Library, ms. XIII A 2; Kolín, 1512 rather than a simple ‘interest’ or ‘disinterest’ in the Agnus chants as such that stood behind all these decisions. At least some shifts within the repertory can be recognized CZ-Pn XIII E 8 as actions performed to make space for another genre. Considering the contents of the Prague, National Museum Library, ms. XIII E 8; origin unknown (OSA?), ca. 1400 Utraquist graduals, the question ‘which repertory’ seems to be easy to answer: strophic CZ-Pst KVš 376 Latin songs (cantiones), the leading poetic genre in the fifteenth century monophonic Prague, Civic State Archive, ms. KVš 376, Bohemia, ca. 1460. repertory. Indeed, some strophic songs found their place in the mass liturgy already CZ-Pu V H 11 in the late fourteenth century, where they were used as tropes to the alleluia or as the Prague, National Library of the Czech Republic, ms. V H 11; Prague, University, ca. 1380 mass ordinary tropes. Even if there is no direct evidence that they would technically CZ-Pu XIV A 1 replace any of the standard chant forms before the first half of the fifteenth century, we Prague, National Library of the Czech Republic, ms. XIV A 1; Kouřim, 1470 should seriously consider it as an option and another piece of a truly individual chant CZ-VB 42 developments in the late medieval Bohemia. Vyšší Brod, The Monastery Library, ms. H 42; Vyšší Brod, 1410

D-Mbs 23286 München, Bayerische Staatsbibliothek, Clm 23286; Praha, Capitular church of All Saints in Prague Castle, beginning of the fifteenth century.

Bibliography

Černý Jaromír (1966), Soupis hudebních rukopisů muzea v Hradci Králové [List of the musical manuscripts of the Hradec Králové Museum], Praha: Universita Karlova (Miscellanea Musicologica 19).

David, Zdeněk (2003), Finding the Middle Way: The Utraquists’ Liberal Challenge to Rome and Luther, Washington, DC: Woodrow Wilson Center Press - Baltimore-London: Johns Hopkins University Press.

Fischer Kurt von (1964), Neue Quellen zum einstimmigen Ordinariumszyklus des 14. und 15. Jahrhunderts aus Italien in Albert van der Linden ed., Liber amicorum Charles van den Borren, Anvers: Lloyd Anversois, pp. 64-68.

Gozzi Marco ed. (2012), Cantus fractus italiano: un’antologia, Hildesheim-New York: Olms.

236 237 CANTUS PLANUS 2014

Graham Garry (2006), Bohemian and Moravian Graduals 1420–1620, Turnhout: Brepols.

Hiley David (1986), ‘Ordinary of Mass Chants in English, North French and Sicilian Manuscripts’, Journal of the Plainsong & Mediaeval Music Society 9, pp. 1-56.

Horyna Martin (2011), ‘Česká reformace a hudba. Studie o bohoslužebném zpěvu českých nekatolických církví v období 1420–1620’ [The Czech reformation and music. A study of vocal music in worship services of Czech non-catholic churches from 1420 to 1620], Hudební věda 48/1, pp. 5-40.

Hospental Christina (2010), Tropen zum Ordinarium missae in St. Gallen. Untersuchungen zu den Beständen in den Handschriften St. Gallen, Stiftsbibliothek 381, 484, 376, 378, 380 und 382, Bern: Lang. VI Karp Theodore (2005), An Introduction to the Post-Tridentine Mass Proper, Middleton: American Institute Saints offices: of Musicology. new themes and contexts Kiss Gábor (2009), Ordinariums-Gesänge in Mitteleuropa: Repertoire-Übersicht und Melodienkatalog, Kassel: Bärenreiter (Mmmae 6).

Koláček Jan (2008), ‘Repertoár a liturgie sedlčanských graduálů’ [The repertoire and liturgy of the Sedlčany graduals], Hudební věda 45/1-2, pp. 5-24.

Limup http://www.clavmon.cz/limup/ (accessed 25.11.2018)

Orel Dobroslav ed. (1922), Kancionál Franusův z roku 1505 [The Franus Kancionál from 1505], Praha: Nákladem obecne jednoty Cyrilské.

Plocek Václav (1973), Catalogus codicum notis musicis instructorum qui in Bibliotheca publica rei publicae Bohemicae socialisticae in Bibliotheca universitatis Pragensis servantur, 2 vols., Praha: Academiae Scientiarum Bohemoslovacae.

Schildbach Martin (1967), Das einstimmige Agnus Dei und seine handschriftliche Überlieferung: vom 10. bis zum 16. Jahrhundert, Dissertation, Friedrich-Alexander Universität, Erlangen-Nürnberg.

Thannabaur, Peter Josef (1962), Das einstimmige Sanctus der römischen Messe in der handschriftlichen Überlieferung des 11. bis 16. Jahrhunderts, München: Ricke.

Vlhová-Wörner Hana ed. (2006), Repertorium troporum Bohemiae medii aevi 2: Tropi ad Kyrie eleison et Gloria in excelsis deo, Praha: Bärenreiter.

– (2008), Three; Nine; Nine and Three. Forms of Kyrie Tropes in the late tradition in Alexander Andreé and Erica Kihlman eds., Hortus troporum, Stockholm: Stockholms Universitet, pp. 138-151.

– ed. (2010), Repertorium troporum Bohemiae medii aevi 3. Tropi ad Sanctus, Praha: Bärenreiter.

– ed. (2013) Repertorium troporum Bohemiae medii aevi 4. Tropi ad Agnus Dei, Praha: Bärenreiter.

Žůrek Jiří (2009), ‘K diskusi nad sedlčanskými graduály’ [To the discussion on graduals from Sedlčany], Hudební věda 46/1-2, pp. 177-186.

238 239 Morné Bezuidenhout

In search of black swans: a computer-assisted approach to interval-pattern recognition

This paper investigates the use of computer-assisted analysis to explore the criteria that Roman Hankeln has identified as stylistic traits of the music in late medieval saints’ offices (Hankeln 1999, 2000, 2001a, 2001b, 2008, 2009). It is a response to the somewhat negative light that Andrew Hughes (2012, 2, pp. 506-17) has cast upon the limitations of electronic searches for «small details» that could aid analytical investigations of the late medieval office repertory. Hughes’s comments about the complexity of electronic searches of plainchant data are rooted in the methodology that he has developed to facilitate this: a system in which search strings for musical data are symbolic representations of pitches (Hughes 1996, pp. 173-174). His concern about the many permutations that pitch-based searches would entail is valid, especially if one attempts to do this with a word processor, as he suggests (2012, 2, p. 507). The solution to Hughes’s problem lies in a re-orientation not only with regard to the tool that is used for searching, but also the encoding of the data. First of all, one cannot use a word processor to facilitate sophisticated searches. The proper tool for this is a database environment that provides a suitable structure for the categorization of the digital data and the relationships that exist among the different categories. Hence, all of the data discussed below reside as joined tables in a database with appropriate software modules for the manipulation and conversion of the data. Secondly, one can simplify the encoding of search strings by converting the pitch data to interval and interval class patterns, as the following overview of the data encoded in the database illustrates.

Debra Lacoste and Kate Helsen (2011) have developed an effective system to encode plainchant pitch data as alphabetic characters, based on Fabian Weber’s Volpiano font, which allows for searchable and sortable data-strings within a database environment. All of the pitch data referred to below was originally encoded by Helsen for her Ph. D. dissertation (2008) on the great responsories in the early twelfth-century antiphoner F-Pn lat. 12044 from the monastery of Saint-Maur de Fossés. Helsen’s pitch and text data were imported into a new database structure in joined tables representing each of the chants as words, phrases and complete chants, together with joined tables extracted from the Cantus Index of the manuscript. The data were also carefully proofread and all chants that could not be encoded as complete chants due to lacunae or other irregularities were excluded, resulting in a closed data set of 923 responsory responds.

241 CANTUS PLANUS 2014 BEZUIDENHOUT — A COMPUTER-ASSISTED APPROACH TO INTERVAL-PATTERN RECOGNITION

The following example is the pitch data for the words Cantabant sancti from the 4. Three leaps in any direction larger than a second or the same, interrupted by a opening of the first-mode responsory Cantabant sancti canticum novi from the feast of second. This also calls for the identification of two interval types (descending and the Holy Innocents in the database (refer to appendix, table 1, p. 251 for a comparison ascending seconds, and descending and ascending intervals larger than a second). of the Volpiano font pitch alphabet and standard medieval pitch nomenclature): The results of the ordered pitch interval pattern is now mapped onto a set of three cd--dhij--h---hgfghghj--h---3. interval classes where 0 = unisons, 1 = ascending and descending seconds, and 2 = all other interval {0,1,2}. Searches will require patterns consisting of reiterations Double and triple dashes (the Volpiano font equivalent for a space between note heads) of three and more instances of interval class 2, and the patterns 2,1,2,2 and 2,1,2,2. signify syllable breaks and word breaks respectively, and the numeric character ‘3’ Here is another example from the office of Saint Maurus. It is the melody of the word (the Volpiano font equivalent of a single bar line) signifies the end of the phrase. To singulorum in the responsory Sanctus domini Maurus, which opens with the pattern facilitate electronic searches, the pitch data are also represented in various stripped 2,2,1,2: 221211111 (dfcdfgfgfe). formats that exclude direct pitch repetitions, the B flat character (‘i’) and space characters, for example, 5. Z-patterns: this is a complicated pattern with a number of characteristics: the cdhjhgfghghjh. outer interval must span a range larger than a fourth; the figure opens with one or a series of ascending or descending intervals that may not exceed the range of Although the pitch data makes provision for pitch-based searches, it serves primarily the outer interval; then another one or series of ascending intervals follow in the as information that can be converted to interval data in order to facilitate searches for opposite direction, once again not exceeding the outer-interval range; finally, another the interval patterns (marker patterns) that Hankeln has classified as characteristic one or series of ascending or descending intervals in the opposite direction also not style criteria of late medieval office chants. Consider the following definitions of five exceeding the range of the outer interval. Hankeln (2008, p. 175) provides a very of Hankeln’s criteria: detailed description of this pattern. The reader should avoid Hughes’s interpretation and his identification of its appearance in some chant melodies (2012, pp. 515-516). 1. Intervals larger than a fourth: in order to search for these intervals, the pitch data These interval patterns are encoded by mapping the ordered interval classes onto is converted to ordered pitch intervals. The stripped melody example given above, a set of three symbolic interval classes indicating unisons, ascending motion, and is represented as the pattern +1+4+1-1-1-1+1+1-1+1+1-1 (cdhjhgfghghjh). Searches descending motion (-, /, \). Additional coding in the database enables searches to limited to individual elements of the set {-4,-5,+4,+5}, representing the ascending eliminate patterns lying outside of the criteria. Consider the following example from and descending fifth and sixth, will undoubtedly be faster and easier to execute the office of Saint Babolenus. It is the second to penultimate pitches of the word and construct than those based on specific pitch combinations representing these accepit in the responsory Venerandus confessor, opening with an extended z-pattern intervals. spanning the range of a seventh: \\\//\\\ (onmlmnlkh).

2. Scales extending beyond the range of a fourth: this also requires ordered pitch Hughes (2012, 2, pp. 499-506 and 509-510) also raises several concerns about the intervals with searches limited to a number of successions of either +1 or -1. likelihood of arriving at generalisations about the late medieval office repertory. These include the relatively small amount of the repertory that has been transcribed and 3. Two leaps in the same direction, larger than a second: these patterns do not analysed, the uncertainties about earlier adiastematic notation, issues surrounding require ordered pitch intervals, since the style criterion calls for the identification of chronological layering, the differentiation between prose and versified offices, and the two interval types only (ascending intervals larger than a second, and descending relationship between text and music. These are very real concerns, but they are not intervals larger than a second). The results of the ordered pitch intervals described central to the issue at hand, which is the relevance of style criteria. These concerns are above are mapped onto a set of four corresponding interval classes where 0 = unison, set aside here by 1) using the repertory in one particular manuscript as a closed data 1 = ascending and descending seconds, 2 = all other ascending intervals and 3 = set (F-Pn lat.12044); 2) applying the computer-assisted analysis without presumptions all other descending intervals {0,1,2,3}. Hence, searches for the marker patterns are about the apparent chronological layers of the chants; 3) limiting the relationship limited to double successions of either interval classes 2 or 3 in this set (2,2 and 3,3). between text and music to the function of syllables, words and phrases as indices for The following is an example, with two successive patterns of this type taken from the the beginning and ending of their associated musical units. melody of the word merore in the responsory Angelus domini candidissimo from the office of Saint Maurus: 22331111111 (fhkhfghjhghg). The swans in the title of this paper refer to Karl Popper’s oft-quoted statement that

242 243 CANTUS PLANUS 2014 BEZUIDENHOUT — A COMPUTER-ASSISTED APPROACH TO INTERVAL-PATTERN RECOGNITION

«no matter how many instances of white swans we may have observed, this does not theme), «melodic families» (e.g., the entire set of melodic types representing first mode justify the conclusion that all swans are white» (2002, p. 4). It is intended to draw antiphons), «standard formulas» (e.g. the melodic gestures, often phrase openings and attention to the obstacles facing style analytical studies in music. Following Leonard cadences, that are the identifying features of melodic types), and «formulaic system» Meyer, one of the tasks of a music analyst is to provide a set of criteria that identify a (e.g. the salient features shared by the phrases of a melodic type). One cannot apply particular musical style by describing and classifying the characteristic patternings or style criteria derived from the observation of one chant type to other types in a cavalier traits of a musical work, or group of works, and in so-doing, disclosing the constraints fashion. It is crucial to carefully delineate the boundaries within which one applies or limitations that determine the context within which these traits appear (Meyer style analytical criteria. 1989, pp. 3, 8-13). This process is similar to the inductive method of the empirical sciences: accumulated observations of replicated instances of particular musical traits, Example 2. Extract from the responsory Beatus Laurentius clamavit how they interact with one another and their functional contexts become premises for general statements or conclusions about the essential characteristics of a musical style. Popper’s observation may serve as a caution against an underlying vulnerability when an analyst employs the results of empirical observations as premises for new observations. Consider, for example, the precarious world of identifying the composer of an unknown composition, derived from the typical stylistic features of a particular composer’s creative procedures. In this regard, one can recall the fiasco surrounding Consider example 2, the opening of the responsory Beatus Laurentius clamavit from the so-called newly discovered Haydn fortepiano sonatas in the early 1990s when Paul the feast of Saint Lawrence. The feast belongs to the earliest layer of the sanctorale, and Eva Badura Skoda, and Howard Chandler Robbins Landon, identified twelfth- appearing in witnesses of the early Roman liturgy such as the Verona and Gregorian century stylistic imitations as previously unknown additions to Haydn’s oeuvre sacramentaries (Frere 1930, pp. 31, 61, 121), and the melody and text can be traced (Koznin 1993). Coming closer to the type of music that this conference is concerned back to the earliest sources. with, consider example 1.

Example 1. Extract from the gradual Haec dies It is a traditional responsory in which the first line opens and cadences with stylistic markers associated with the later repertory: a) an opening gesture consisting of an interval larger than a second, followed by a second and three intervals larger than a second, and b) a cadential figure containing three successive intervals larger than a second. The use of characteristic initial and cadential patterns, often related to mode and genre, are a well-known phenomenon in traditional Gregorian chant. They function as points of reference or orientation that aid the recall of melody for singers that have to retain a large amount of melodies (Treitler 1974, pp. 353 ff.; Helsen It prominently displays several of the characteristics that Hankeln has identified 2008, pp. 465-467). In Helsen’s and Lászlo Dobszay and Janka Szendrei’s analyses as stylistic markers of late medieval chants for saints’ offices: z-patterns in close of the responsories, patterns exhibiting the interval and pitch content of these two proximity, a scale passage extending beyond a fourth; leap combinations consisting of examples are identified as typical initial and cadential gestures of the first phrase of three (or more) successive intervals larger than a second in any direction, sometimes many traditional second-mode responsories (Helsen 2008, pp. 100-101, 104; Dobszay interrupted by a second, or two intervals larger than a second in the same direction. and Szendrei 2013, 1, p. 71). Example 2 is thus a good example of standard formulas However, this example is from neither a late medieval nor an office chant. It is the last belonging to the same musical types, melodic type, and melodic families functioning phrase of the response of the traditional Gregorian Easter gradual: Haec dies. At this in this fashion. point, it may be useful to recall Willi Apel’s advice that each single plainchant in the Gregorian repertory «belongs to a liturgical class from which it receives the general Example 2 does not undermine Hankeln’s identification of interval patterns, such as characteristics of musical form and style that set it wholly apart from an item belonging this, as markers of the later repertory. He has, from the outset, stressed that they can to a different class» (Apel 1958, p. 304). In a similar vein, Leo Treitler (1974, pp. 347- appear in traditional Gregorian melodies and that there are two features making them 353 and 357-362) differentiates between «musical types» (antiphons, responsories, significant in the later melodies: a higher rate of occurrence, and their uncharacteristic graduals, etc.), «melodic types» (e.g. first mode antiphons based on the same Gevaert position or placement within a melody (Hankeln 2001a, p. 159). In this responsory,

244 245 CANTUS PLANUS 2014 BEZUIDENHOUT — A COMPUTER-ASSISTED APPROACH TO INTERVAL-PATTERN RECOGNITION the stylistic markers have a low rate of occurrence and when they appear they lie in markers appearing at the structurally significant positions. The cadential patterns of structurally significant positions. both phrases consist of combinations of z-patterns and leaps (karo suo and dominus) and the second phrase concludes with a descending scale run exceeding a fourth Examples 3 and 4 contain extracts from the feast of saint John the Evangelist. The feast (within a z-pattern) and a sub-tonal cadence. But the question one has to ask here is also belongs to an early layer of the Sanctorale (Frere 1930, pp. 32, 60, 84-85, 143-144), if they are functioning within a traditional context. The conclusion of the first phrase but the two responsories are stylistically different. would have been more significant as a traditional marker if it appeared in a different context. Recall example 3 where this same pattern appeared as part of the phrase Example 3. Extract from the responsory Qui vicerit faciam illum preceding the repetenda in its traditional context. In Apparuit it appears in the ‘wrong’ place and lacks the typical opening and closing gestures that surround its counterpart in the previous example. In short, here it appears outside of its traditional context, while retaining its function as a closing gesture. The cadence of the second phrase is not a traditional marker. Although Helsen (2008, pp. 74-76) classifies a number of first mode phrases with cadences ending with this sub-tonal cadence (CDD) as a standard element (d2), it is very unstable whenever it appears, retaining the sub-tonal cadence as its most characteristic common feature. A survey of the first mode chants Example 3 is the cadential phrase preceding the repetenda from the respond of the in which this standard elements appears (with the aid of Helsen’s data files), suggests responsory Qui vicerit faciam illum. The responsory consists entirely of traditional an association with the later repertory. It is absent from the earliest antiphoners (Albi, material, but as in the previous example, this example contains an intervallic pattern Compiègne, Hartker and Mont-Renaud) and it appears in chants that exhibit a high characteristic of the later repertory: a z-pattern from the concluding pitch of the first density of late medieval markers or feature other characteristics of the later repertory. syllable of nomen to the penultimate pitch of the second syllable of nomen. Several studies of the second-mode responsories have pointed out that the entire cadential The examples that I have offered are too limited to lay the basis for a strong hypothesis formula is similar to the closing formula of the traditional second-mode responsory about the constraints determining the interaction of the interval patterns and their verse tone and that it typically precedes the repetenda (Helsen 2008, pp. 104-106; functional contexts. However, they offer a point of departure for further investigations Dobszai anad Szendrai 2013, 1, p. 72). In this position it has a very particular function: along these lines. They provide some confirmation for Hankeln’s comment about the as the conclusion of the first part of the respond and the conclusion of the verse it presence of late style markers in the traditional and the later repertory. In the former, signifies the place in the chant were the repetenda begins, during the response and they appear in a limited fashion and when they do, they tend to appear in functionally after the verse. important strategic positions such as, for example, the openings and cadences of Example 4. Extract from the responsory Apparuit karo suo Iohanni the phrases in a chant. In the later repertory their density increases, they continue functioning as structural markers, but also appear outside of their traditional context. This has consequences for analytical methodologies aiming to quantify stylistic criteria. Since the significance of the patterns in the later chant repertory lies not only in their density but also in their positioning in non-traditional contexts, their appearance in traditional contexts in this repertory is insignificant as style markers and should be excluded from the quantitative results.

Example 4 is the opening phrases of the responsory Apparuit karo suo Iohanni. This This section presents the results of computer-assisted quantitative analyses of the chant does not appear in extant sources written before the end of the first millennium responsory responds for the feast of Saint Babolenus in F-Pn lat. 12044. This feast is and it bears some of the hallmarks of later chants added to the traditional repertory. In without doubt a later addition to the office repertory, which is confirmed not only by addition to the higher density of the Hankeln’s marker patterns and a newly composed its absence from the earlier manuscript from the same monastery, F-Pn lat. 12584 verse, the entire melody also exhibits features that David Hiley has highlighted (2004), (manuscript F in Corpus antiphonalium officii), but also by a reading of the chant such as an excessive range, sub-tonal cadences, and the use of the final and fifth melodies that takes style criteria associated with the later repertory into consideration. above the final as melodic goal pitches. As in the previous example, there are stylistic Table 2 shows the analytical results.

246 247 CANTUS PLANUS 2014 BEZUIDENHOUT — A COMPUTER-ASSISTED APPROACH TO INTERVAL-PATTERN RECOGNITION

Table 2. Analysis of responsories for the feast of Saint Babolenus 7. Only patterns lying within the elements (phrases) that Helsen delineates are counted. Intervals between these components are considered as dead intervals. Incipit P Ind Mode G Pitch A Ind S-T Cnt The other columns are: the ‘Mode’ of the chant with ‘S’ indicating a newly composed Post gloriosa proelia 9.83 1S 100 11 2 109 responsory verse; ‘G[oal] Pitch’, the percentage of polysyllabic words in each chant Confessoris Christi 4.52 2S 55.56 11 2 119 ending on the final, octave above the final, the fourth below the final and the fifth Beatum confessorem Babolenum 3.99 3S 58.33 11 0 119 above the final; ‘A[mbitus] Ind[ex]’ – ‘1’ on the left shows a range extending below the theoretical octave range of a mode, ‘1’ on the right shows a range extending above Inclyte pater monachorum 7.8 6 4S 46.15 01 1 114 this range; ‘S[ub]-T[onal cadence]’ the number of these cadences in the chant; and Divinis insultans operibus 3.99 5S 81.82 11 0 104 ‘C[ou]nt’, the length of the chant, excluding intervals between phrases, expressed as Pro salute gregis hujus 5.84 6S 91.67 11 0 102 the of the sum of these intervals. Venerandus confessor amoris 10.00 7S 66.67 11 1 173 These results allow us to return to Popper’s warning about the perils of predictions Gloriosus pater Babolenus 7.6 8 8S 73.33 11 1 171 based on inductive methods and earlier observations in this paper that style criteria for Inter venerandos domini 6.37 1S 88.24 11 3 149 a group of works are not only determined by their occurrence but also by the context Insignis pater Babolene 3.10 1S 75 11 3 77 within which they appear. One of the clearest signs that, in the case of Babolenus, we are considering later additions to the traditional repertory is the modal numerical Pastor noster venerande tibi 5.84 3S 60 10 0 123 ordering of the first eight chants. This an undisputed index for the later repertory, Sanctus domini confessor 8.99 1S 76.19 11 3 185 but, as Hughes has pointed out, it is a feature that one should treat with some care when it does not appear, since re-orderings sometimes took place when offices were transferred from a monastic to a secular environment (Hughes 1983, pp. 30, 31, 38, The second column (‘P[attern] Ind[ex]’) presents a score weighted according to the passim). Hence, a feast without modal ordering may be a later version of one with length (the total number of intervals) of a chant. Each chant is awarded a score out of modal ordering. Another index for the later repertory is the newly composed verses ten with the highest score (ten/ten) awarded to the chant with the highest density of instead of the traditional responsory verse tones. However, even this feature should marker interval patterns. This score is also awarded in relation to all of the responsories sometimes be treated with care. Walter Frere mentions the possibility of traditional in the sanctorale of F-Pn lat. 12044. The following criteria determine the identification responsory responds with newly composed verses (1901, p. 5), and Apel suggests of the marker interval patterns: that in some of these cases, the verses might date from a «period in which the system of responsorial tones was not yet fully established and standardized» (1958, p. 241). 1. Only leap combinations, intervals larger than a fourth, scale patterns exceeding the range of a fourth, sub-tonal cadences and z-patterns are counted. One should also proceed with caution when considering the significance of sub- tonal cadences, the concentration on goal pitches and excessive range, three well- 2. In the case of leap combinations and large intervals, subsets are excluded. Hence documented features of the later repertory. None of these features can be treated in a fifth, or any other pattern, contained within larger patterns, is only counted once. isolation, nor do low counts necessarily signify a traditional chant. In the introduction 3. In the case of scales, only the largest superset of one or more smaller subsets is to the edition of the Historia sancti Ludgeri, I have suggested that there might be a counted. relationship between modal affiliation, and the presence of sub-tonal cadences, goal- pitch concentration and excessive melodic range. In this investigation, data from a 4. Z-patterns are treated in isolation, that is, z-patterns that are subsets of leap small group of offices were collated, which showed that chants most likely to have combinations are not ignored. lower counts of one or more of these markers belong mainly to the third, fourth, and fifth modes (Bezuidenhout 2010, pp.xl -xliii). Although the dataset underlying this 5. In the case of z-patterns, only the largest superset of one or more smaller subsets speculation requires expansion, the data third-, fourth-, and fifth-mode chants in is counted. the Babolenus office provide further support for this surmise. 6. Patterns containing unison repetitions are ignored.

248 249 CANTUS PLANUS 2014 BEZUIDENHOUT — A COMPUTER-ASSISTED APPROACH TO INTERVAL-PATTERN RECOGNITION

Finally, a consideration of the pattern score. Here, it would be useful to have an idea of Appendix some elementary statistics. The dataset contains 441 chants from the sanctorale that contain one or more of the marker patterns. One of the Babolenus chants (Venerandus Table 1. Volpiano and medieval pitch nomenclature confessor amoris) is the chant with the highest score of marker patterns. However, the Babolenus chant Insignis pater Babolene (example 5) lies slightly below the average 9 a b c d e f g h j k l m n o p q r s score of 3.29 for all of the sanctorale chants. Γ A B C D E F G a b c d e f g aa bb cc dd Example 5. The respond of the responsory Insignis pater Babolene i = B-flat accidental

Bibliography

Apel Willi (1958), Gregorian chant, Bloomington: Indiana University Press.

Bezuidenhout Morné (2010), Historia sancti Ludgeri, Ottawa: The Institute of Medieval Music (Music Musicological Studies 65/15).

Dobszay László - Szendrai Janka eds. (2013), Responsories, 3 vols., Budapest: Balassi Kiadó.

Frere Walter Howard (1901-1924), Antiphonale sarisburiense: A reproduction in facsimile from early Although one could argue that since this is the shortest chant in the feast, the length of manuscripts with a dissertation and analytical index, 6 vols., London: Gregg. chant could influence the pattern index, a brief survey of the data for chants of other – (1930), Studies in early Roman liturgy, 3 vols., Oxford: Oxford University Press, vol. 1: The calendar feasts indicates that this is not always the case. The style of this chant can only be (Alcuin Club Collections 28). assessed by considering all of the style criteria pointing to its later origin. Even though Hankeln Roman (1999), Die Antiphonen des Dionysius-Offiziums in Clm 14872 (St Emmeram, XVI. Jh.) it has a relatively low pattern count, it has a high goal-pitch rate, a wide range, a newly in Walter Berschin - David Hiley eds., Die Offizien des Mittelalters: Dichtung und Musik, Tutzing: composed verse, a restless melodic dynamic (see especially the long final melisma), Schneider, pp. 109-128. and, within its short length of five phrases, three sub-tonal cadences. – (2000), In der Werkstatt des Cantors: Hinweise zu einigen Merkmalen mittelalterlichen Komponierens in Martin Angerer ed., Ratisbona - Die königliche Stadt: Neue Forschungen zum mittelalterlichen Regensburg, Regensburg: Universitätsverlag Regensburg, pp. 32-45. In conclusion, while computer-assisted searches are useful tools for the identification of chants that contain certain interval patterns and other style criteria, searches of this – (2001a), Antiphonen süddeutscher Heiligen-Offizien des Hochmittelalters in László Dobszay ed., Cantus th nature can only function as an aid to the analyst. They provide a point of departure Planus: papers read at the 9 meeting, Esztergom & Visegrád 1998, Budapest: Hungarian Institute of Sciences, pp. 151-172. for investigations that would require insight and knowledge of the chant repertory and literature. Armed with this knowledge, an analyst will take care not to assume – (2001b), ‘«Properization» formal and changes in high medieval saints’ offices: the offices for saints Henry and Kunigunde of Bamberg’, Plainsong and Medieval Music 10, pp. 3-22. that criteria derived from the known repertory will necessarily be relevant for future investigations of the unknown. Furthermore, the evidence of the investigated repertory – (2008), Old and new in medieval chant: finding methods of investigating an unknown region in Ole should serve as a warning that style criteria cannot be considered in isolation from Kongsted - Niels Krabbe - Michael Kube - Morted Michelsen eds., A due: musical essays in honour of John D. Bergsagel & Heinrich W. Schwab, Copenhagen: The Royal Library, pp. 161-180. one another. The contexts within which they appear are overarching criteria that help to define the place of a chant within a specific chronological layer. And finally, there – (2009), ‘Zur musikstilistischen Einordnung mittelalterlicher Heiligenoffizien’, Etudes grégoriennes 36: Lingua mea calamus scribæ: mélanges offerts à madame Marie-Noël Colette, pp. 147-157. should be awareness that the currently defined criteria are not a closed set: additional criteria may be added and existing ones may be redefined. Helsen Kate (2008), The great responsories of the Divine Office: Aspects of structure and transmission, Dissertation, Universität Regensburg http://epub.uni-regensburg.de/10769/ (accessed January 10, 2016).

Helsen Kate - Lacoste Debra (2011), ‘A report on the encoding of melodic incipits in the Cantus database with the music font «Volpiano»’, Plainsong and Medieval Music 20, pp. 51-65.

250 251 CANTUS PLANUS 2014

Dirk van Betteray Hiley David (2004), Early cycles of office chants for the feast of Mary Magdalene in John Haines - Randall Rosenfeld eds., Music and medieval manuscripts: essays dedicated to Andrew Hughes, Ashgate: Aldershot, pp. 369-399. Textinterpretation und Formelkomposition

Hughes Andrew (1983), ‘Modal order and disorder in the rhymed office’, Musica disciplina 37, pp. 29-52. Heiligenoffizien im Codex Hartker

– (1996), Late medieval liturgical offices: Resources for electronic research: Sources and chants, Toronto: Pontifical Institute of Mediaeval Studies (Subsidia Mediaevalia 24).

– (2012), The versified office, 2 vols., Lions Bay: The Institute of Medieval Music (Musicological Studies 971). In der Forschung gibt es mittlerweile zahlreiche Untersuchungen aus musikwissen­ ­ schaftlicher und theologischer Sicht, die einen Zusammenhang herstellen konnten Koznin Allan (1993), ‘Discovered sonatas may be faked Hadyn’, The New York Times, December 28. zwischen patristischer Schriftauslegung­ und kompositorischer Text­ausdeutung im Meyer Leonard (1989), Style and music: Theory, history and ideology, Philadelphia: University of gregorianischen Choral.1 Der vorliegende Beitrag beschäftigt sich speziell mit dem Pennsylvania Press. Wort-Tonverhältnis in formel­haften Kompositionsstilen von Offiziums­antiphonen. Popper Karl (2002), The logic of scientific discovery, London: Routledge, or. ed. Logik der Forschung, Wien: Durch Vergleich von Formelvarianten­ an konkreten­ Beispielen aus Heiligenoffizien­ Spinger 1935. des sogenannten Codex Hartker2 werden Möglich­keiten und Grenzen vor­gestellt,

Treitler Leo (1974), ‘Homer and Gregory: The transmission of epic poetry and plainchant’, The Musical inwieweit die tatsäch­lich verwendete­ Formel aufgrund des zu Grunde liegenden Quarterly 60, pp. 333-372. Textes gewählt worden sein könnte. Außerdem­ wird die konkrete Neumierung von formel­haften Wendungen­ daraufhin­ untersucht,­ inwieweit sie beispielsweise durch Neumen­zusätze (Buchstaben, Episemata, Liqueszenzen) eine Formel an den kon­ kreten Text und seine Aussage anpasst. Die Ergebnisse­ aus dem Codex Hartker werden mit Parallelstellen­ aus der jüngeren Sankt Galler Schwesterhand­ ­schrift G3883 und aus dem diastematischen­ Codex E6114 verglichen, der aus dem nahen Einsiedeln­ stammt. Dadurch können ebenfalls Rückschlüsse­ gewonnen werden, inwie­weit tatsächlich­ in den Heiligen­offizien des Codex Hartker Neumen­zusätze auf­grund des Textes vorge­ nommen wurden und ob diese in späterer Zeit auch noch inten­diert waren.

Zu Beginn der Unter­suchung soll ein Fall­beispiel stehen, bei dem gleich mehrere Parameter­ in die gleiche Richtung weisen, dass hier melodische Formeln aufgrund­ der Textinter­ ­pretation verän­ ­dert bzw. angepasst­ wurden (Tabelle 1, p. 254). Es handelt­ sich um den Text in monte sancto. Dieser kommt im Codex Hartker jeweils in zwei direkt hinter­einan­der fol­gen­den Antiphonen­ der ersten Commune-Nokturn für «einen Märtyrer oder Bekenner» (G391,178) vor sowie wenige Seiten später an entsprechen­ ­ der Stelle für «einen Priester oder Bekenner» (G391, 182). In der jeweils ersten Anti- phon (Praedicans bzw. Beatus iste) handelt es sich um den sechsten Vers aus Psalm 2 (Ps. 2,6): constitutus est in monte sancto eius (er wurde ein­ge­setzt auf seinem hei­ligen Berg). In der jeweils zweiten Antiphon (Voce mea bzw. Tu es gloria) ist es der fünfte

1. Theologische Untersuchungen, zum Beispiel: Lumma 2006; Riede 2010; musikwissenschaftliche Untersuchungen, zum Beispiel Praßl 2004; van Betteray 2007; Praßl 2008.

2. CH-SGs 390/391 um 1000 (von nun an: G390 bzw. G391). Zur Entstehungsgeschichte des Codex Hartker siehe Pouderoijen (2009). Die drei Hauptneumatoren unterscheiden sich nicht im Gebrauch der Funktionalität der Neumen trotz kleiner Unterschiede im Schreibstil und in einigen kontextuellen Interpretationen.

3. CH-SGs 388 (von nun an: G388) aus dem zwölften Jahrhundert.

4. CH-E 611 (von nun an: E611) aus dem vierzehnten Jahrhundert.

252 253 CANTUS PLANUS 2014 VAN BETTERAY — HEILIGENOFFIZIEN IM CODEX HARTKER

Tabelle 1 Vers aus Psalm 3 (Ps. 3,5): et exaudivit me de monte sancto suo (und er erhörte mich von seinem heili­gen Berg); bzw. im zweiten Bei­spiel in direkter Anrede: et exaudisti Praedicans Tr Tr P Tr Tr Tr C Tr nkPliq C Tr Tr G391,178 me de monte sancto tuo. (= G388, 371) III Praedicans sancto NR1 [35] G G G A G G G A G E G Aliq G F E E In der Antiphon erhält die Akzentsilbe­ von große Bedeutung­ durch Hochton und einen nicht kurrenten Pes, der noch zusätzlich durch eine Liqueszenz er- con- sti- tu- tus est in mon- te sanc- to e- ius weitert wird. Gleichzeitig wird monte durch eine kurrente Clivis eher entlastet, so dass E611, 255 G G D F G A G F E E sancto wie zu erwarten im Binnenverhältnis dieser beiden Worte das wichtigere ist. Alternative2 G G F G F E E Genauso verhält es sich auch in der Antiphon Voce mea. Wieso aber wird sancto in der Antiphon Praedicans durch eine nicht kurrente Neume und zusätzliche Liqueszenz Voce mea Ve V V Tr V V V Tr V V V Tr Tr G391, 178 so stark hervorgehoben, wo es doch eine viel einfachere Formelalternative gegeben (= G388, 371) VII hätte, die dem Binnenverhältnis von monte und sancto ebenfalls entsprochen hätte? NR [36] C C C A C H A G A C H G G Und warum ist in der Antiphon Voce mea die Akzentsilbe von sancto nicht liqueszent, obwohl es doch ein solche Alternative gegeben hätte? Bevor wir diese Frage beant- et ex- au- di- vit me de mon- te sanct- to su- o worten, schauen wir auf das andere Antiphonenpaar Beatus iste und Tu es gloria mea E611, 259v C C H A H C A G E A H A G G (Tabelle 2). Alternative3 d C H C dCliq A G G Hier finden wir die Beobachtungen des ersten Antiphonenpaars bestätigt. Bei Beatus 1. NR= Nocturnale Romanum 2002. 2. NR 166 Antiphon Orietur; vgl. ebenfalls E611, 17v. iste ist die Akzentsilbe von sancto wiederum liqueszent, obwohl es auch hier eine li- 3. NR 72* Antiphon Agathes laetissima, vgl. ebenfalls G390, 121. queszenzlose Alternative gegeben hätte. In der Antiphon Tu es gloria verhält es sich Abkürzungsverzeichnis in Neumentabellen, hier und weiter unten siehe p. 264 parallel zum ersten Antiphonenpaar wiederum umgekehrt. Außerdem erhält sancto in der Antiphon Beatus iste wiederum eine besonders starke Betonung – hier durch einen

Tabelle 2 Pes auf der Endsilbe von monte, der für einen vorbereitenden Stau sorgt.

Beatus iste Tr V V V V V V P Vliq C Tr Tr Warum erfährt monte sancto im Kontext von Psalm 2 in beiden Fällen eine andere Be- G391, 182 handlung als im Kontext von Psalm 3? Gerade die Parallelität zweier Antiphonenpaare (=G388, 375) I ist hier auffällig, von denen jeweils die beiden Antiphonen eines Paares im Offizium NR [84] C D F G F F F G A F Eliq G F D D nach Hartker in unmittelbarer Nähe stehen. Eine mögliche5 Begründung findet sich con- sti- tu- tus est in mon- te sanc- to e- ius in den Psalmenkommentaren des heiligen Augustinus. Diese patristischen Psalmen- E611, 262v F F G A F Dliq F D D kommentare waren den Sankt Galler Mönchen zur Zeit der Niederschrift des Codex Hartker bekannt und geläufig.6 In diesen Psalmenkommentaren findet sich tatsächlich Alternative1 F F E F G F E D D eine je andere Auslegung für monte sancto in Psalm 2 und in Psalm 3. Tu es gloria Tr V Ve V V V V V C P V Tr Tr G391, 182 Für Psalm 2 heißt es: «Ergo iste sensus est: Ego autem constitutus sum rex ab eo super (=G388, 375) 7 VIII ecclesiam sanctam eius, quam montem appellat propter eminentiam et firmitatem» NR [86] F F A C C H C A G F G A A G G (Dies aber ist der Sinn: Ich bin nämlich von ihm zum König eingesetzt über seine heilige Kirche, die er Berg nennt wegen ihrer Erhabenheit und Festigkeit). et ex- au- di- sti me de mon- te sanct- to tu- o

E611, 263 F G A C C C C A G F G A A G G 5. Vortragstexte, zu denen auch die Antiphonen des gregorianischen Chorals gehören, sind prinzipiell immer mehrdeutig. Sonst wären verschiedene stimmige künstlerische Interpretationen eines Gedichtes oder Musikstückes a priori nicht 2 Alternative A G A Cliq H G G möglich.

1. Vgl. Riede (2010, pp. 7, 23). 6. In der Sankt Gallen Klosterbibliothek liegen eigene Abschriften der augustinischen Psalmenkommentare aus dem neunten Jahrhundert vor (CH-SGs 162-166). 2. Vgl. NR 71 Antiphon Si ignem bzw. G390, 121. Die drittletzte Silbe ist potentiell liqueszierbar. In der Antiphon Si ignem fehlen jedoch die phonetischen Voraussetzungen. 7. Augustinus 1990, 1, 5.

254 255 CANTUS PLANUS 2014 VAN BETTERAY — HEILIGENOFFIZIEN IM CODEX HARTKER

Für Psalm 3 heißt es hingegen: «De memetipso, tamquam de monte sancto suo, me nicht kurrente Neumen (auf retibus) bezogen werden können auf eine Predigt Papst exaudivit»8 (Von sich selbst aus, sowie von seinem heiligen Berg, hat er mich erhört). Gregors des Großen, der in seiner Homilie «habita ad populum in basilica beati Andreae apostoli, in die natalis eius» (Gehalten vor dem Volk in der Basilika des seligen Apostels Während also in Psalm 2 mit monte sancto nach Augustinus die Kirche gemeint ist, be- Andreas, an seinem Geburtsfest),11 also an berufenem Ort und zu berufener Zeit sagt: deutet dieser Ausdruck in Psalm 3 Gottes Wohnstatt bzw. Gott selbst. Die letztere Be- deutung entspricht dem Wortsinn, zumal Gott im Kontext ohnehin die handelnde Per- Ad vocem dominicam uterque iste piscator quid aut quantum dimisit, qui pene nihil habuit? […] son ist. Hier bedarf es keiner besonderen Vertonung. In den auf Psalm 2 basierenden Aestimationem quippe pretii non habet: sed tamen regnum Dei tantum valet, quantum habes. Valuit Petro et Andreae dimissis retibus et navi. Antiphonen gibt es aber eine weitere Bedeutung zusätzlich zu und hinter dem Wort- Was oder wie viel haben denn auf den Ruf des Herrn hin jene beiden Fischer aufgegeben, die doch sinn. Die besondere Vertonung, gerade auch die Verwendung der Liqueszenzen lassen so gut wie nichts besaßen? […] Es gibt zwar keinen festen Preisanschlag, doch kostet das Reich diese für den Sänger und Hörer deutlich werden.9 Wir finden in den hier untersuchten Gottes so viel, wie du besitzt. Es kostete Petrus und Andreas den Verzicht auf Netz und Boot.12 Antiphonenpaaren also zahlreiche Indizien, die darauf hinweisen, dass die tatsächliche Vertonung aufgrund patristischer Textinterpretation gewählt wurde, da es aus rein musi- Nicht zuletzt aufgrund der großen liturgischen Nähe von Vigilmesse und erster Nok- kalischer Sicht ebenbürtige Alternativen gegeben hätte. Die jüngere Neumenhandschrift turn liegt es nahe, auch die nicht kurrenten Clives in der hier vorliegenden Offizium- G388 übernimmt diese Intentionen noch, in der diastematischen Quelle E611 finden santiphon als ein retardierendes Element zu erklären, das aufmerksam machen soll wir neben den üblichen Varianten ostfränkischen Choraldialekts durchgängig eine eher auf die hinter dem reinen Wortsinn liegende patristische Interpretation. Schließlich 13 aus innermusikalischen Gründen gewählte Melodik, die grammatikalische Wortakzente wäre formelgemäß auch eine Fassung ohne Episemata möglich gewesen. aber durchaus berücksichtigt. Somit lässt sich an diesem Beispiel auch zeigen, wie sich interpretatorische Sichtweisen am selben Entstehungsort im Laufe der Zeit verändern.10 Gezeigt werden sollen nun einige weitere Beispiele, in denen eine Melodieformel bewusst verändert wurde, um den Textsinn an der entsprechenden Stelle zu verdeut- Auch die Antiphon Relictis retibus (G391, 161) aus der ersten Nokturn des Andreas­ lichen. Michael Riede hat im Jahr 2010 an der Universität Wien eine Untersuchung festes (30. November) zeigt in ihrer Vertonung Spuren patristischer Theologie, die in über die Finalformeln des Protus authenticus im Codex Hartker vorgelegt. Riede der jüngeren Handschriften G388 wohl nicht mehr verstanden wurden oder nicht kommt dabei zu dem Schluss, dass am Schema der Formel festgehalten wird, «wenn mehr gewollt waren. Nur der Codex Hartker notiert nämlich auf der ersten Silbe und es die Textsituation erlaubt», dass aber das «Schema bereitwillig verlassen wird», der letzten Silbe dieses Beispiels eine nicht kurrente Clivis (Tabelle 3). «wenn es die Textsituation erfordert». Wie hier schon beobachtet,14 stellt auch Riede fest: «Spätere Handschriften weichen manchmal von den Sonderfällen Hartkers ab Tabelle 3 und greifen auf Formeln streng nach Schema zurück» (2010, p. 92).

G391, 161 Ce V V Cl Tr Ce Riede identifiziert in seiner Untersuchung des Codex Hartker fünf verschiedene Final- G388, 354 C V V C Tr C cento-Formeln für den Protus authenticus und führt diese direkt auf unterschiedliche sprachklangliche Gegebenheiten wie Wortakzente und Wortlängen zurück. Dass die NR 2* D C F G F E D D C musikalische Gestalt von Antiphonen Rücksicht auf die Textur nimmt, verwundert Relictis retibus re- lic- tis re- ti- bus nicht. Auch ist bekannt,15 dass die Schöpfer gregorianischer Antiphonen mehrfach

E611, 153 D C F F G E D D C die Textstellung veränderten, um eine bessere Verbindung von sprachlichen und mu- sikalischen Betonungen herzustellen. Besonders interessant sind aber die Fälle, wo scheinbar keine der möglichen Formeln ausgereicht hat für eine ideale Vertonung des zugrunde liegenden Textes. Für die Communio der Vigilmesse des Andreasfestes Venite post me mit der gleichen Textgrundlage (Mt, 4,20) konnte bereits an anderer Stelle das Wort retibus betreffend dargestellt werden (van Betteray 2007, pp. 148-150), dass Episemata, Liqueszenz und 11. Gregorius Magnus, Homilia v, 1997, pp. 102-103. Nach Fiedrowicz (1997, p. 102) wird die Homilie auf den 30.11.590 datiert, also das Andreasfest.

12. Gregorius Magnus, Homilia v, 1997, pp. 104-105.

8. Augustinus 1990, 1, 9. 13. vgl. z.B. Antiphon Sunt de hic stantibus (G390, 64).

9. Zur Bedeutung der Liqueszenz für die Wortinterpretation siehe van Betteray (2007, pp. 195-199). 14. Vgl. auch van Betteray (2014).

10. Vgl. auch Praßl (2011). 15. Siehe Rumphorst (1992).

256 257 CANTUS PLANUS 2014 VAN BETTERAY — HEILIGENOFFIZIEN IM CODEX HARTKER

Beginnen wir mit der Antiphon Ecclesiae pastores (G391, 29) aus dem Gallus-Officium Die Wichtigkeit der Worte os und viri lässt sich durch Episemata an den Clives (16. Oktober). Dass dieses Offizium des Patrons und Namengebers in Sankt Gallen mit hervor­heben. Eine melodische Änderung ist dafür nicht notwendig. Gerade aber am besonderer Feierlichkeit begangen wurde, zeigt schon das Dedikationsbild des Codex Ursprungsort der Verehrung des Heiligen Gallus, wo man sozusagen auf den Heiligen Hartker, auf dem zu sehen ist, wie der Mönch Hartker dem Heiligen Gallus seinen Co- zeigen kann, scheint auch das letzte Wort istius besonders bedeutend zu sein: Der dex überreicht.16 Außerdem schmücken zwei große Zierinitialen das Gallus-Offizium. Heilige Geist hat heute wirklich gesprochen durch den Mund «dieses Menschen da». Diese besondere Betonung entsteht dadurch, dass die Endsilbe von viri mit einem Die gesamte Antiphon Ecclesiae pastores zeigt in ihrer Neumierung durch den Codex F statt einem D vertont ist. Der Quartsprung an ungewöhnlicher Stelle schafft Auf- Hartker die Hervorhebung zentraler Worte. Im ersten Teil finden sich retardierende merksamkeit für das Kommende. Das F an dieser Stelle bezeugen auch E611 sowie und nicht kurrente Neumen auf pastores und doctrinam. Zusätzlich wird audientes weitere Quellen auch außerhalb der Sankt Galler Lokaltradition.17 Es hat sich über die mit einem Betonungspes hervorgehoben und die Akzentsilbe von Galli durch eine Jahrhunderte erhalten, obwohl jüngere Quellen prinzipiell eher dazu neigen, schema- Liqueszenz verlängert: tische Formeln beizubehalten.18

Die Hirten (pastores) der Kirche hören (audientes) die Lehre (doctrinam) des Heiligen Die Magnificat-Antiphon Astiterunt für das Fest Johannes und Paulus (26. Juni) weist (Gallus). Der zweite Antiphonenteil enthält die Reaktion der pastores ecclesiae und ebenfalls eine Formelveränderung auf (Tabelle 5). damit die Lehre für das Kirchenvolk: «Vere spiritus sanctus hodie locutus est per os viri istius» (Wahrhaftig hat der Heilige Geist heute gesprochen durch den Mund dieses Tabelle 5 gerechten Mannes). Die Hauptbetonung liegt gleich im ersten Wort des zweiten Teils: vere (wahrhaftig) – neumiert mit einem nicht kurrenten Torculus. Dann folgen wei- Text et ami- ci de- i fa- cti sunt tere zentrale Begriffe, die nicht kurrente Neumen enthalten, nämlich spiritus (Geist, G391, 85 Tr, Tr, Tr V nkP Pes subbipunctis V Pe V Pes quadratus) und sanctus (heilig, Pes quadratus). Es ist also wahrhaftig der Heilige Geist, der hier gesprochen hat, so die Aussage der Antiphon, und zwar «durch den Melodie nach G391 F F F F E F F G F E D C D D Mund dieses Mannes» (per os viri istius). Istius bedeutet «dieser da», der in meiner Melodie E611, 189 F F F F E F G F E D C D D Nähe steht, auf den ich mit dem Finger zeigen kann. Genau das trifft auf die Mönche Melodie nach Schema 1 F F F E F G F E D C D D in Sankt Gallen in Bezug auf ihren Namensgeber und Ortspatron in vieler Hinsicht zu.

1. Vgl. z.B. Antiphon Positis autem genibus die Stelle ne status illis hoc peccatum (Fest des heiligen Stephanus, G390, p. 59). Nach diesen Vorbefunden ist der Schluss dieser Antiphon per os viri istius für die hier Nach Riede Finalcento protus de tertio (2010, p. 33). Oder auch die Antiphon Qui enim corpori (Fest des heiligen Stephanus, G390, 59). vorliegende Untersuchung besonders interessant, denn hier wird die melodische Formel um der Textaussage willen verändert. Die Neumierung des Codex Hartker verfeinert diese Aussage weiter (Tabelle 4). Die Formel wurde hier verlassen, um das Wort die zu betonen. Von den Heiligen heißt Tabelle 4 es hier: «calicem Domini biberunt et amici Dei facti sunt» (sie haben den Kelch des Herrn getrunken und sind Freunde Gottes geworden). Eine formelgemäße Vertonung Text per os vi- ri is- ti- us versähe den letzten Wortakzent mit einer Betonung. Die Clivis davor wäre eine nicht betonte Durchgangsgruppe. Bei unserem Beispiel entstünde eine Betonung auf facti G391, 129 nkP Ce Ce V O Tr Tr sunt, die dem Sinnzusammenhang entgegen steht: Die Heiligen sind nicht Freunde E611, 234 FG FE DC F E D D Gottes «geworden», sondern, indem sie den Kelch des Herrn getrunken haben, sind sie

Melodieschema1 FG FE DC D FE D D 17. Z.B. die Sankt Lambrechter Quelle A-Gu 30 aus dem vierzehnten Jahrhundert oder die Augsburger Quelle (Druck) DK-Kk 3449 80 XII aus dem Jahr 1580. 1. vgl. z.B. Antiphon Ecce veniet deus et homo die Stelle throno alleluia (G390, 21). 18. Unterstrichen wird die Intention der Betonung des Wortes istius auch durch den Oriscus auf der Anfangssilbe is-, da so die praetonische Silbe deutlicher von der Akzentsilbe getrennt wird. Der Oriscus legt auch eher einen Tonsprung F-D statt des Tonschritts E-D nah. Dieser lässt sich allerdings in den untersuchten diastematischen Quellen nicht belegen, was wiederum am ostfränkischen Choraldialekt liegen könnte, aus dessen Verbreitungsraum diese stammen. In Sankt Galler Kodizes lassen sich zwar ebenfalls Spuren dieses Choraldialektes finden, aber in so frühen Quellen wie dem Codex Hartker 16. Es ist zu bemerken, dass die Namen auf dem Bild später hinzugefügt wurden. Siehe Pouderoijen (2009, p. 85). sind diese bei weitem weniger ausgeprägt als in jüngeren Quellen.

258 259 CANTUS PLANUS 2014 VAN BETTERAY — HEILIGENOFFIZIEN IM CODEX HARTKER

Freunde «Gottes» geworden. Die melodische Formel wurde also aufgrund der Textaus- Tabelle 7 sage bewusst verlassen bzw. variiert. Im nachfolgenden Kontext (Tabelle 6) hingegen entspricht das Betonungsschema der Formel: Dominus na- tus est in ter- ris E611, 22 Bb A G G F A G Tabelle 6 G390, 55 V V Tr Cep nkPliq Tr Text il- lis hoc pec- catum Stephanus nas- ce- re- tur in cae- lis G390, 59 V nkP V C cel Tr, Tr E611, 22 F D C D G E F D D Melodie nach Schema F E F G F E D D so auch E611, 24v G390, 55 V C cel nkP Tr Cep Tr Tr

In diesem Ausschnitt aus der Antiphon Positis autem genibus vom Fest des heiligen Tabelle 8 Stephanus (26. Dezember) heißt es: «Domine ne statuas illis hoc ad peccatum» (Herr, Melodie D DC F G FA A rechne ihnen dies nicht als Sünde an). Der Wortakzent auf peccatum ist auch der letzte AM1 / E611 Sinnakzent, der durch die mit einer Clivis mit cel(eriter) vertonten leichten Neume Neumen nach G390/391 V C V V nkP Ve vorher seine notwendige Betonung erhält, wie es diese Formel vorsieht. G391, 175, Comm. plur. Mart. Ein interessanter Vergleich (Tabelle 7) bietet sich im Offizium des Protomärtyrers Ste- G388, 367 Cor- po- ra sanc- to- rum AM 648 phanus (26. Dezember). Noch vor dem Apostel und Evangelisten Johannes, der Christi Botschaft in die Welt hinausträgt und verkündet, gedenkt die kirchliche Liturgie dem G391, 181, Comm. unius Mart. Qui mi- hi mi- ni- strat G388, 373 391/388 391 ersten Märtyrer Stephanus. Überschwänglich heißt es in der ersten Antiphon der ers- AM 640 Tr Ce ten Nokturn (G390, 55): «Hesterna die Dominus natus est in terris ut hodie Stephanus G390, 112, Agnetis Stat a dex- tris e- ius nasceretur in caelis» (Gestern ist der Herr auf Erden geboren worden, damit heute G388, 114 390/388 390 Stephanus im Himmel geboren wird). Hier wird der Geburt des Herrn (Dominus) auf AM 798 (Stans) Tr Ccel Erden die Geburt des heiligen Stephanus im Himmel gegenübergestellt.19 Die Verto- G390, 64, Joannis Evang. Ec- ce pu- er me- us nungen der Worte in terris und in caelis sind zwar nicht gleich. Sie stehen ja auch G388, 69 390/388 390/388 AM 256 Ccel Cep an unterschiedlichen Stellen der Spannungskurve einer kurzen, aus einem Satz be- G391, 186, Comm. unius Conf. Eu- ge ser- ve bo- ne stehenden Antiphon. Beide Präpositionen in sind jedoch mit einer liqueszenten Neu- G388, 379 391/388 391 me, dem Cephalicus, versehen, der alleine durch seine klangliche Mehrwertigkeit im AM 670 Cep V Gegensatz zu einer Virga schon retardierend wirkt und so das folgende Wort durch G391, 114, Nativitas Mariae Na- ti- vi- tas tu- a Verzögerung hervorhebt.20 G388, 312 391 AM 1035 Ce In den folgenden Beispielen von Antiphonenanfängen im Protus authenticus wird die G391, 108, Decoll. Jo. Bapt. Pu- el- lae sal- tan- ti melodische Formel durch Neumenzusätze wie Episemata, Liqueszenzen, oder Buch- G388, 298 391/388 AM 1026 Cep staben an die Textaussage angepasst (Tabelle 8). 1. AM = Antiphonale Monasticum 1934.

19. Dieselbe Aussageabsicht lässt sich auch im ersten Responsorium prolixum der ersten Nokturn Hesterna die beobachten, das einen fast identischen Text vertont; vgl. hierzu van Betteray (2014 pp. 34- 36).

20. Die Antiphon Ecce dedi verba (G391, 80) zeigt am Ende beispielsweise, dass die Virga vor dem Terzsprung in einer Diese in vielen Offiziumsantiphonen des Protus authenticus (erster Modus) zu fin- Schlussformel auch nicht liqueszent erscheinen kann trotz potentieller Liqueszierbarkeit. Dass in einer Schlussformel in der Mehrzahl der Fälle an dieser Stelle eine potentiell liquesziebare Silbe auch tatsächlich liqueszent erscheint, hat dende Intonationsformel beginnt mit der Finalis (D), erhält durch eine Clivis mit auch damit zu tun, dass das Betonungs- und Akzentgefüge in der Formel immer gleich ist. Diese Formel wird ja eben zur der Untersekund (DC) Schwung und erreicht über die Terz (F) den Tenor (A). Die Betonung eines solchen Textes verwendet, dessen Aussage sie unterstreichen kann. Diese Situation spricht also nicht gegen das Vorhandensein eines engen Text-Melodie-Verhältnisses. Hauptbetonung dieser Formel liegt auf der vorletzten Silbe, auf welcher der Tenor

260 261 CANTUS PLANUS 2014 VAN BETTERAY — HEILIGENOFFIZIEN IM CODEX HARTKER erreicht wird. Die Neumierung mit einer nicht kurrenten Neume, nämlich einem In den anderen beiden Fällen machen die Episemata gerade auf kontextuell beson- Pes quadratus, unterstreicht diese Betonung innerhalb der Formel, die jedoch ders starke Betonungen aufmerksam: Bei Nativitas tua ist nativitas das wichtigere gleich­zeitig als Intonationsformel zum Folgenden hin öffnet. Die hier ausgewähl- Wort der beiden. Es wird aber durch die melodische Formel benachteiligt. Die nicht ten Beispiele aus Heiligenoffizien unterscheiden sich erwartungsgemäß nicht von kurrente Clivis verleiht dem Wort den notwendigen Nachdruck. der Verwendung dieser Formel im übrigen Repertoire. Auch diese Beispiele zeigen, dass die Clivis auf der zweiten Silbe im Normalfall kurrent ist, um die Bewegung in Bemerkenswert ist ebenfalls der andere Fall eines Episems auf der Clivis in der An- Schwung zu bringen. Das wird bei Stat a dextris eius durch cel(eriter) verdeutlicht, tiphon Qui mihi ministrat (Joh, 12,26): Der Text lautet nach Hartker komplett: «Qui damit das einsilbige a und dessen wenn auch leichter Wortakzent nicht retardierend mihi ministrat me sequatur et ubi ego sum ibi sit et minister meus» (Wer mir dient, wirkt. Eine ähnliche Funktion hat das cel(eriter) bei Ecce puer meus: Hier wird folge mir nach; und wo ich bin, dort wird mein Diener auch sein). Eigentlich müsste durch den Zusatzbuch­staben eine Abgrenzung des Ausrufs Ecce vom weiteren Text ministrat im Verhältnis zu mihi das wichtigere Wort sein. So war es den Menschen verhindert, die den Sinn der Aussage entstellen würde. Durch eine solche Abgren- der Zeit Jesu und auch den mittelalterlichen Menschen aus ihrer Alltagswelt be- zung könnte die Aussage nämlich heißen: «Seht her! Mein Knecht!», im Sinne von: kannt: Ein Diener folgt seinem Herrn auf Schritt und Tritt. Dem Wort mihi wird hier «Bin ich nicht außergewöhnlich, dass ich mir einen solch prächtigen Knecht leisten durch das Episem eine größere Bedeutung geschenkt, weil Jesus seinen Jüngern kann». Die Aussage der Antiphon nach Jesaja 42, 1f. geht aber tatsächlich folgen- etwas Anderes verheißt: «ubi ego sum ibi sit et minister meus» (überall wo ich bin, dermaßen weiter: «Ecce puer meus electus, quem elegi: posui super eum spiritum dort wird mein Diener auch schon sein). Jesus verheißt seinen Jüngern hier nicht meum» (Seht, das ist mein erwählter Knecht, den ich erwählt habe. Ich habe meinen weniger als die Aufnahme in das Reich seines Vaters.22 Wenn sich aber die Nachfolge Geist auf ihn gelegt).21 Jesu von dem Dienst für einen anderen Herren grundlegend unterscheidet, dann ist nachvollziehbar, warum das Episem an der Clivis mihi betont. Die Handschrift Zwei der untersuchten Stellen weisen ein Episem an der Clivis auf der zweiten Silbe G388 zeigt auch bei diesen Fällen im Vergleich, dass in späterer Zeit wohl eher rein der Formel auf: Qui mihi ministrat und Nativitas tua. In beiden Fällen steht dort melodische Parameter wie zum Beispiel eine beständige Formel stärker in den Blick eine Akzentsilbe, die durch eine nicht kurrente Neume das ihr zustehende Gewicht der Notatoren gerieten: In G388 weist die Clivis in Qui mihi ministrat und Nativitas erhält. Bei Puellae saltanti fällt die Clivis ebenfalls mit der Akzentsilbe zusammen. tua nämlich kein Episem auf. Die Grundformel bleibt bestehen. Andererseits muss Hier aber wird die Clivis diminutiv liquesziert. Im Vergleich zeigt sich auch in die- die Beibehaltung der Liqueszenz bei Puellae nicht unbedingt als Ausdruck der Text- sem Fall, wie durch unterschiedliche Neumenzusätze eine an sich für diese Textur ausdeutung ausgelegt werden. Eine Liqueszenz kann in jüngeren Quellen häufig passende Melodieformel enger an den konkreten Text und seine Aussage angepasst aus rein phonetischen Gründen stehen, falls sie nicht einfach beim Abschreiben wird. Das Wort puellae ist auf seiner Akzentsilbe aufgrund des phonetischen Be- übernommen wurde.23 Ein Vergleich mit der diastematischen Quelle E611 zeigt in fundes zwar grundsätzlich liqueszierbar. Doch scheint eine diminutive Liqueszenz allen hier untersuchten Beispielen, dass die Melodik dieser Intonationsformel un- gerade auf einer Akzentsilbe nicht sinnvoll zu sein, da sie die Betonungsqualität der verändert geblieben ist. Akzentsilbe schmälert. Im Textzusammenhang wird aber deutlich, dass das Wort puellae das am wenigsten wichtige im gesamten Kontext ist: «Puellae saltanti impe- Alle diese Beispiele zeigen, dass die jeweilige konkrete Melodie- und Neumenfas- ravit mater: Nihil alius petas, nisi caput Joannis» (Dem tanzenden Mädchen befahl sung nicht nur den zugrunde liegenden Text und seine grammatikalische Struktur die Mutter: Verlange nichts anderes als das Haupt des Johannes). Die diminutive berücksichtigt, sondern darüber hinaus die Interpretation des Textes verdeutlicht. Liqueszenz verhindert hier also eine dem Textsinn nicht entsprechende Betonung. Die Anzahl der Indizien und deren Häufung im Hinblick auf dasselbe Ziel hin lassen diese über das Maß des Zufalls hinausgehen. Die zugrunde liegende Textinterpre- tation scheint dabei keine beliebige zu sein, sondern sich auf Grundlagenwerke pa- tristischer Theologie zu beziehen, die für die Schreiber und Hörer der Entstehungs- 21. Die Neumierung unterstreicht diese Interpretationsthese nicht nur durch ein cel(eriter) auf Ecce. Zusätzlich findet sich zwischen meus und electus noch ein «x» (= exspectate), das die Weiterführung auf das nach meus Folgende zeit der Quellen größte Autorität besaßen. Gerade die Untersuchung formelhafter genauso verdeutlicht wie die nicht kurrente Clivis auf e-lectus und ein weiterer liqueszenter Pes quadratus auf elegi. Diese Betonungssilbe wird schließlich abgefangen durch einen Tractulus mit Episem. Auch die Liqueszenz auf der Endsilbe von puer gehört in diesen Zusammenhang. Das Ganze bildet ein stimmiges Konzept, durch rhythmische Varianten den 22. Das macht das Wortpaar illic - ubi deutlich, das erklärt auch den Betonungspes im Codex Hartker auf ibi. Und eben Sinn dieses Textes innerhalb einer formelhaften Melodik zu verdeutlichen, die selbstverständlich der Metrik des Textes diesen Sachverhalt hat Luther in seiner Bibelübersetzung deutlicher herausgestellt als die Autoren der Einheitsübersetzung, entspricht. Ein einziges dieser Indizien kann Zufall sein, aber eine Häufung wie im vorliegenden Fall scheint schon auf eine indem er das Wort «auch» nachstellte: «Vnd wo ich bin / da sol mein Diener auch sein» (1545), zitiert nach: (16.06.2014) im Gegensatz zu «Und wo ich bin, dort wird auch mein Diener sein» (Einheitsübersetzung der Bibel, noch bewusst gewesen zu sein, übernimmt er doch das cel(eriter) auf der Clivis, obwohl diese Handschrift generell nur zitiert nach: http://www.bibelwerk.de/Bibel.12790.html/Einheits%C3%BCbersetzung+online.12798.html (16.06.2014). sehr wenige Zusatzbuchstaben notiert. Bei Stat a dextris eius beispielsweise wird das cel(eriter) nicht übernommen (vgl. Tabelle 8). 23. Vgl. u.a. van Betteray (2007).

262 263 CANTUS PLANUS 2014 VAN BETTERAY — HEILIGENOFFIZIEN IM CODEX HARTKER

Kompositionen war erhellend, weil es für diese Fälle jeweils authentische rhyth- Bibliography misch-melodische Formelalternativen gibt, gegen die sich der Schreiber jedoch in Einzelfallen aufgrund des textlich-melodischen Zusammenhangs entschieden zu Augustinus Aurelius (19902), Enarrationes in Psalmos, Eligius Dekkers - Johannes Fraipont eds., 3 vols., haben scheint. Das nimmt den Alternativen einen rein hypothetischen Charakter Turnholt: Brepols (Corpus Christianorum, Series Latina 38-40). und macht sie zu einem wertvollen Negativindiz für die dargestellte These. Die Beschränkung auf die Untersuchung von Heiligenoffizien konnte exemplarisch ver- Betteray Dirk van (2007), Quomodo cantabimus canticum Domini in terra aliena: Liqueszenzen als Schlüssel zur Textinterpretation; eine semilogische Untersuchung an Sankt Galler Quelle, Hildesheim-Zürich: deutlichen, dass die Schreiber durchaus die konkrete Situation vor Ort zur Zeit der Olms (Studien und Materialien zur Musikwissenschaft 45) Niederschrift mit im Blick hatten. Die jüngeren Vergleichshandschriften zeigten, – (2014), ‘Hic est vere Martyr, qui pro Christi nomine sanguinem suum fudit. Representations and wie sich Intentionen tradieren oder auch durch die Jahrhunderte ändern können. reflections of violence and suffering in the Responsoria prolixa of saints’ offices in Codex Hartker’, Die bisherigen Forschungsergebnisse auf dem Gebiet des Wort-Tonverhältnisses Plainsong and Medieval Music 23/1, pp. 31-50. im gregorianischen Choral konnten somit auf Antiphonen der Heiligenoffizien im Dobszay László - Szendrei Janka eds. (1999), Antiphonen, Mmmae 5. Codex Hartker übertragen und an Formelkompositionen auf eigene Weise verifiziert werden. Fiedrowicz Michael ed. (1997), Gregor der Große, Homiliae in Evangelia, 2 vols., (1, 1997; 2, 1998), Freiburg im Bresgau: Herder (Fontes Christiani 28).

Gregorius Magnus, Homiliae, see Fiedrowicz 1997.

Iversen Gunilla - Bell Nicolas eds. (2009), Sapientia et eloquentia. Meaning and Function in Liturgical Poetry, Music, Drama, and Biblical Commentary in the Middle Ages Abkürzungsverzeichnis Neumentabellen , Turnhout: Brepols).

Lumma Liborius Olaf (2006), Qui manducat carnem meam et bibit sanguinem meum. Theologische C Clivis Implikationen der Gregorianischen Communio-Antiphonen de evangelio im Messproprium des cel celeriter Temporale, Wien-Münster: Lit (Liturgica Oenipontana 5). Cep Cephalicus Pouderoijen Kees - de Loos, Ike (2009), ‘Wer ist Hartker? Die Entstehung des Hartkerischen Antiphonars’, e mit Episem Beiträge zur Gregorianik 45, pp. 41-56. liq mit Liqueszenz nkP nicht kurrenter Pes Praßl Franz Karl (2004), ‘Scriptor Interpres. Von Neumenschreibern und ihren Eigenheiten’,­ Beiträge zur O Oriscus Gregorianik 37, pp. 55-72. P Pes – (2008), ‘Gregorianische Gesänge als Zeugnisse für patristisches Schriftverständnis’, Beiträge zur Tr Tractulus Gregorianik 45, pp. 41-56. V Virga – (2011), ‘Sankt Galler Handschriften als Ausdruck konkreter Aufführungstraditionen’, Beiträge zur Gregorianik 52, pp. 89-110.

Riede Michael (2010), Die Finalformeln des protus authenticus im Codex Hartker. Eine centologische Untersuchung, Universität Wien (online: http://othes.univie.ac.at/9992/).

Rumphorst Heinrich (1992), ‘Gesangstext und Textquelle im Gregorianischen Choral’, Beiträge zur Gregorianik 13/14, pp. 181-209.

264 265 Roman Hankeln

The articulation of direct speech in responsories

Much has been written about the importance of the text for the theory, composition and practice of plainchant.1 Undisputed is the formal structural relationship between both media: larger or minor syntactical units such as half and full sentences acted obviously as the main landmarks medieval composers used for the structuring of musical phrases.2 Much more debated is the question of whether or not the musical observance went any further and included also reflections of the meaning and the emotions articulated in the texts. Isidore of Seville had demanded that the informed lector should understand the grammatical structure and theological meaning of Scripture, which would enable him to «control the technique of oral delivery (vim pronunciationis) without impediment, in order that he may move the minds and feelings (sensus) of all to understand by distinguishing between the kinds of delivery, and by expressing the feelings (affectus) of the sententiae».3 In a comparable vein, the Commemoratio brevis places the text at the centre of its attention («meruimus uerba maiestatis sue [id est: Domini] in os sumere»), demanding high-quality musical performances based on understanding and skill («scienter et ornate confiteamur nomini sancto eius»), where «heart» and «voice» are combined in order to move people to holy affection («[…] si scilicet et animo apud Deum dulciter canitur, et homines canoris dulcedo sanctu affectu commouet»).4 If we take them seriously, these and other indications suggest an interest in meaning and emotion that may have led to more than mere formal parallels between music and text. Today, we are however unable to verify whether meaning and emotion really received a musical articulation in the chants and how that was accomplished. This is mainly due to the fact that a comprehensive, systematic and differentiated study of this question that would take the various genres and the historical stratification of the plainchant repertory into account, is lacking. The present study proposes to take a new initiative in this field. It takes its starting point in the observation that there exist text-situations that require a specific musical response, at least from a well informed singer/composer. Among the most obvious of these situations is, for example, the enjambement in poetic texts. In the

1. See for example Schlager (2000).

2. By referring to the elementary grammar of Donatus (fourth century A.D.), Johannes Cotto gives an explicit parallelisation of musical and linguistic phrases and sub-phrases. See Babb and Palisca (1978, p. 116).

3. Isidore (De ecclesiasticis officiis) translated by Parkes (1993, p. 35). In a similar vein, Amalar of Metz demanded that the sense of what was sung in the responsory verse should find a sensible continuation when the repetendum-section was repeated. This indicates that the message of the chant texts was not a side issue. Amalar does however not speak directly about musical matters in this context. See Hanssens (1948, p. 362, l. 34–36).

4. See Bailey (1979, p. 26, 28).

267 CANTUS PLANUS 2014 HANKELN — THE ARTICULATION OF DIRECT SPEECH IN RESPONSORIES same way questions, exclamations, and the change of speakers in dialogues provoke the Example 2. R. Inclitus Oswaldus, Saint Oswald-historia. GB-Cmc F.4.10, fol. 260v singer/composer to take a musical stand on the form, content, and emotional attitudes R. Inclitus Osuualdus, domini bello moribundus, ... of the text. The present contribution focuses on only one of these text-situations: the phenomenon of (introduced) direct speech. The following two examples show some of its characteristics and why it has been chosen for this case study.

In the churches of medieval Western Christendom, during matins on Good Friday, the seventh-mode responsory Tenebrae factae sunt was sung. The text of Tenebrae is based on a passage in the gospels of Matthew and Mark which reports on the solar eclipse during Christ’s passion (Matthew 27:45-46; Mark 15:33-34). This account includes Jesus’s question from the cross: «Deus meus, Deus meus, ut quid dereliquisti me?» (which is a quotation of the beginning of Psalm 21). Compare phrases 6 and 7 in example 1: In both pieces the direct speech of the protagonists is introduced by a short Example 1. R. Tenebrae factae sunt, Cao 7760. F-Pn lat. 12044, fol. 96v characterising sentence. Musically, both introduction and direct speech are R. Tenebre facte sunt, dum crucifixissent Hiesum Iudei, et circa horam nonam … separated from another through a clear caesura. The relationship between both phrases is characterised by a contrast in register. In the second example, the direct speech also shows a much greater amount of melismatic behaviour.

The two examples we have studied so far show the linguistic phenomenon of direct speech in its fullest form: here, a personal utterance in the indicative mood is introduced by a short, more or less complete sentence that informs us about the persons speaking, the situation they are in, as well as the character of the subsequent direct speech (example 1, phrases 4 and 5, «exclamavit Hiesus uoce magna»; example 2, phrase 3, … «clamat in extremum»). A common feature of these introductory sentences is that they …*et inclinato capite, emisit spiritum. Tunc unus ex militibus […] are neutral in tone. I therefore call them ‘neutral units’, or ‘n-units’. We should bear in mind that direct speech that is linked to one or several n-unit(s) confronts us with at In the preceding phrases 4 and 5, Christ’s words are introduced by a short sentence least two speakers and usually with different tones of speech. which characterises them as an outcry: «exclamavit Hiesus uoce magna». The It seems unclear how the transition between n-unit and direct speech was actually introduction ends with a caesura on the note F. In phrase 6, Christ’s direct speech starts read during the Middle Ages. No philological study on this question is known to me. a sixth higher, on d. Soon afterwards, the final’s upper octave g is touched. Phrases 6 The examples in Malcom B. Parkes’ overview of medieval punctuation (1993) indicate and 7 are positioned in the top register of the gamut. that the transition was regularly denoted as a major or minor pause.5 In Peter Wagner’s In the eighth matins responsory from the eleventh-century office in honour of Saint examples of punctuation signs, and of the tones for prayers and lessons, we encounter Oswald we hear the king’s last words during the fatal battle at Maserfelth. Already most often the punctus circumflexus, and the punctus between n-unit and direct fatally wounded, Oswald prays for the souls of his fellow soldiers: «Miserere salus speech. These signs denote pauses after the smallest syntactical incision, the comma, animarum!», phrase 4 in example 2. and after the full sentence.6 The melodic inflections of these incisions usually have a melodic downward movement.7 Musically, Oswald’s prayer is set as an extended melisma over ‘miserere’, which thrice touches the highest note of this eighth-mode chant, d. A zigzag gesture at the start of 5. Compare here the various kinds of punctuation indicating the pause between the n-unit and the direct speech in the following passage (modern transcription): «malediceret his verbis: ‘maledictus Chanaan puer’ […]», in Parkes (1993, p. 278, phrase 4 traverses the whole ambitus of an octave in just eight notes. As in R. Tenebrae plate 65, l. 9; p. 280, plate 66, col. 1, l. 28 f.; p. 282, plate 67, col. 1, paragraph 2, l. 3; p. 284, plate 68, col. 2, l. 1 f.).

(example 1), Oswald’s direct speech is prepared by an introduction in phrase 3 which 6. Punctus circumflexus, see Wagner (1912, p. 86, par. 2, l. 2): «Dicit enim scriptura [sign] Omnis […]»; ibidem l. 11: «Isaias characterises the subsequent direct speech as an outcry: «clamat in extremum: […]». In autem dixit [sign] Domine […]». Punctus, see Wagner (1912, p. 86, last paragraph, l. 2 ): «Hec dicit dominus. Propter […]». this case, the introduction concludes with a caesura on the final. The beginning of the 7. The punctus elevatus which marks the end of the colon has in the lesson-tones a rising melodic movement (Wagner (1912, p. 89). In Wagner’s examples this sign was only very rarely used to mark the end of the n-unit; see Wagner (1921, p. direct speech starts in phrase 4 a fourth higher on c, again a clear contrast in register. 42) for the example Haec dicit Dominus.

268 269 CANTUS PLANUS 2014 HANKELN — THE ARTICULATION OF DIRECT SPEECH IN RESPONSORIES

Direct speech is not an unknown phenomenon in plainchant scholarship, but with the Group 1 (forty-one pieces of 231): the pieces of this group place the direct speech in exception of some shorter remarks, no musicological study has provided a systematic the B-part, the repetendum. The complete A-part functions as an introduction to the description of the phenomenon.8 The intention of this contribution is to report on the direct speech, as for example in: first results of some exploratory steps I have undertaken in that direction. This was A motivated by the impression that the phenomenon of direct speech might play a not n Ingressus angelus ad Mariam ait: B insignificant methodological role for the investigation of music and text relationship DS *«Ave Maria, gratia plena, Dominus tecum. V in medieval plainchant. To this end, I will discuss the setting of direct speech in DS Benedicta tu in mulieribus, et benedictus fructus ventris tui». B responsories from the Franco-Roman corpus and the high and later medieval DS *«Ave […] (Cao 6963) repertory of historiae. Of course, the antiphons also contain direct speech, but for Group 2 (forty-four pieces of 231): in this group the A-part again starts with an n-unit. obvious reasons, I had to concentrate on that genre of the office repertory which, But unlike the previous example, the first sentence of the direct speech begins not with due to its greater length and formal and stylistic scope, offers a higher probability of part B but already in the A-part. The direct speech is continued in the B-part: encountering diverse and complex solutions to this textual problem. A n Cum vidisset beatus Andreas crucem, exclamavit et dixit: One starting point for the classification of the responsory texts was their edition A DS «Salve crux, quamdiu fatigaris, in the fourth volume of Hesbert’s Corpus Antiphonalium Officii (1970 – henceforth B DS *exspectans me sicut exspectasti Dominum et Magistrum meum. Cao), both in the temporale as well as the sanctorale. In the Cao, 1921 responsory- V DS O bona crux, diu desiderata, et jam concupiscenti animo praeparata». texts are available, and 463 of these represent introduced direct speech. Only 231 B DS *«Exspectans […]» (Cao 6378) musical settings of these texts were available to me in modern editions, however.9 The high and later-medieval repertory of local, non-Cao-responsories from the The second sentence of Andrew’s direct speech covers the verse. In the B-part, a portion historiae-repertory, available to me in modern transcriptions, comprised 868 pieces of the apostle’s first sentence will then be repeated («expectans me […]»). in total.10 Out of this total, only sixty responsories showed the full form of direct Group 3 (thirty-two pieces of 231). These responsories place their direct speech speech. The following contribution, thus, is based on the analysis of 291 responsory exclusively in the verse, most frequently in the second part of the verse. The first part melodies (231 Cao, plus sixty non-Cao-pieces). is used for the introduction:

Direct speech and responsory-structure A n Cantabant sancti canticum novum ante sedem Dei et Agni, In order to describe the relationship between direct speech and musical form in the B n *et resonabat terra in voces illorum. responsories, I created a list which registered the frequency and the placement of V n: Sub throno Dei omnes sancti clamant: the segments of direct speech and the n-units with respect to the main parts of the V DS «Vindica sanguinem nostrum, Deus noster!» responsory: the first part of the response – short ‘A’ in the following – the second B n *Et resonabat […] (Cao 6266) part of the response, the repetendum – short ‘B’ – and the Verse – short ‘V’. (In the following, ‘n’ and ‘DS’ signify n-units and segments of direct speech.) On this field I A special, enigmatic phenomenon of text-display can be observed only in the repertory observed a multitude of various combinations which were sorted into groups of the of non-Cao-responsories. In a small group of this repertory (only five of sixty pieces) same structural type. Here I can present only examples for the largest groups found the end of the respond is an explicit introduction to the direct speech that is displayed in the Cao. The later repertory shows approximately the same distribution. in the verse. The text of the respond, thus, lacks a proper semantic conclusion. See this example from the office in honour of Saint Iacobus (Compostela):

A 8. The late Andrew Hughes (2011, vol. 1, pp. 215–218) quoted a number of examples from the offices in honour of Saint n Huic Jacobo condoluit dominus tempore passionis sue, velut karus karo suo Edmund and Saint Francis. Most relevant for the present discussion has been the short chapter xxxi in Johner (1953, pp. 446–454). See also the discussion of the Tenebrae-responsory by Maurus Pfaff in Fellerer (1972, pp. 258–259). mesticiam carnis sue B 9. See Dobszay and Szendrei (2013). Useful was also the transcription of the responds of F-Pn. lat. 12044 in Helsen (2008). n [*] ostendens ei et dicens: V 10. For the later, non-Cao responsory texts and melodies I relied on transcriptions of locally transmitted saints’ offices DS «Tristis est anima mea usque ad mortem […]». which were made during the 1990s at the University of Regensburg (see http://www.uni-regensburg.de/Fakultaeten/phil_ B n [*] Ostendens [ei et dicens:]11 Fak_I/Musikwissenschaft/cantus/index.htm ‘Offices of the Saints’), the volumes of the Historiae-series, as well as the older editions of important saints’ offices as that of Thomas Becket (ed. Slocum 2004), Saint Francis of Assisi (ed. Felder 1901), or Charlemagne (ed. Jammers 1934). 11. See the trancription in Wagner (1931, p. 72).

270 271 CANTUS PLANUS 2014 HANKELN — THE ARTICULATION OF DIRECT SPEECH IN RESPONSORIES

This text arrangement leads to a contradiction between form and content. The final Example 4. R. Audit in evangelio, Franciscus-historia, ed. Felder (1901, p. xxiv) cadence of the respond sets a musical conclusion where the syntax and sense of the respond’s text, the n-unit, is not complete. All five pieces of this group show repetendum- cues after the verse. This indicates that the scribes expected at least formally a return to the repetendum which implies that the respond would be concluded with the semantically incomplete end of the n-unit.12

Musical settings of direct speech In the Cao-responds I counted 181 longer or shorter musical segments with direct The choice of a single word as indicator of direct speech looks plausible especially speech texts. These segments were introduced or followed by shorter or longer n-units. in poetry, which operates with rhyme, syllable count and alternating accentuation, Of the direct speech segments, forty-one percent (seventy-five of 181), exhibit clearly all formal features which restrict the space for linguistic manoeuvre in the context contrasting musical behaviour when compared to that of the surrounding n-units. of direct speech. Like the dicit dominus-interjections, the inquit-units are not sharply This means that these direct speech-segments are usually situated in a higher register separated from the surrounding melodic context. Again this is probably due to their than the surrounding segments, and that they usually make a fresh melodic start at subordinate syntactic status. their beginning. In the later repertory, approximately the same intensity of musical articulation of direct speech can be observed. Most segments of direct speech in the Cao- Emotion and rhetorical emphasis responds (163 of 181) are introduced using n-units. The endings of these n-units were It is a truism that direct speech can bring a considerable element of realism, even nearly always marked with a clear musical caesura on the final (ninety-one percent, drama into the texts. When we hear the saints, the angels, even God, Jesus and 148 of 163 cases). For this, almost always downward melodic movements were used. Mary speak with their own voice, we become part of the ‘real action’. For the singers This observation also holds true for the later responsories. The introductory n-unit of these chants, the matter seems even more intense: they take the speech of these is thus normally not concluded with a half cadence on the upper fifth for example, figures literally into their own mouth. Sometimes we even seem to get a glimpse but with a caesura on the final. This treatment seems to be in line with the melodic of the feelings of these figures in their speeches, as we have seen in the two first downward movements in connection with the comma and the periodus I mentioned examples. Both responsories have exclamatory texts which are articulated in clearly with regard to the tones of lessons and prayers. A small group of Cao-responsories dramatic contexts. In example 1, this might even have led to an exceptional choice (twenty-three pieces of 231), features the use of n-units not as introductions but as where its formulaic structure is concerned. In Helsen’s study of the formulaic short interjections or additions. In melodic terms, these are only weakly separated structure of responsories the end of the n-unit, phrase 5, is labelled ‘f’. Phrase 6, from the music of direct speech. Frequently the biblical phrase ‘dicit dominus’ is used, Christ’s direct speech with the label ‘l4’. Helsen’s table of the seventh-mode formulae as shown in example 3. shows that the specific combination of these formulae occurs only in this special case.13 Can we interpret this choice as an additional indication of the exceptional Example 3. R. Iuravi, Cao 7045. F-Pn lat. 12044, fol. 5v nature of this situation, and possibly as an attempt to convey the emotionality of its content musically? Examples 1 and 2 seem to show an obvious connection between an emotional text and a highly contrasting musical setting which recalls similar phenomena of text setting, for example from the Renaissance. Interesting as they may be, such pieces are not very numerous, however. In order to put them in the right perspective in relation to the use of direct speech in the responsories in general, I must at this point mention representatives of another group, which shows In the later repertory, the n-unit may only be marked with the word «inquit», as in the opposite text-music relationship (example 5, p. 274). example 4 from the rhymed office in honour of Saint Francis:

13. Helsen 2008; see Appendix/Chapter_3_6_Mode_7/Database_2_Responds_horizontal. 12. Here the remaining pieces of this group: Kunigunde (Bamberg): R. Cvmque deo dilecta […] ait ad dominum: V. «Iudica The list shows that the seventh mode formulae which follow a preceeding element ending on F are ‘l1’, ‘l2’ and ‘l8’. In the iudicium meum […]»; transcription: Hankeln 2010, pp. 93-114, 107; Dionysius (Regensburg): R. Cvm dominus Christus Ihesus responsories of F-Pn. lat. 12044 none of these connections shows however the direct leap of a sixth between the elements. […] ait Dionisius: V. «Hec nox […]»; transcription: Hankeln 1998, p. 8; Cancius et socii (Aquileia): R. Orabant claris vocibus […] The ‘l4’-formula is often connected to a preceding formula which ends on G (labelled ‘g’, ‘g1’, or ‘g4’), so that direct leaps hanc orationem: V. «O Ihesu Christe domine […]»; transcription: Snoj 2003, pp. 47-48); Oswald (Northumbria): R. Pasche die […] of a fifth (G-d) between the end of G-element and the start of ‘l4’ can be observerved rather frequently. This is also true for reddit Aydanus benedictionem, que pernament in hunc diem: V. «Numquam marcescat […]»; transcription: Huseby 2014, p. 18. the connection ‘g1, 3, 4, or 5’ and ‘l2’.

272 273 CANTUS PLANUS 2014 HANKELN — THE ARTICULATION OF DIRECT SPEECH IN RESPONSORIES

Example 5. R. Cumque omnia, Mariae ad Nives-historia. D-Mbs, clm 9508, fol. 192v characterising the speaker.15 Peter’s exhortation is a perfect example of this principle.

R. Cumque omnia Liberius seriatim exponeret, que sibi et Iohanni Patricio fuerant revelata, … The apostle’s own calm words show the strength of his faith. Here we can hardly expect a special musical emphasis of the direct speech – at least not out of emotional reasons. But let us take a closer look at the music: the beginning of Peter’s direct speech in phrase 6 shows a rising gesture which is intensified through a short leap combination, C-E-G. The start of the direct speech, Peter’s imperative «nolite», is thus clearly articulated. … The contrast between phrase 6 and the preceding phrase is however not very strong in musical terms. Peter’s direct speech in phrase 6 is not situated in a completely different register, and it also shares some melodic material with the preceding phrase. As a whole, the apostle’s speech thus does not seem particularly exposed musically. This musical representation of medium intensity corresponds to the meaning and the attitude of Peter’s calm and comforting words. In the repertory, we encounter examples of this kind quite frequently. My general impression was in fact that of a prevailing rather plausible In this example the verb of the n-unit in the preceeding phrase characterises the relationship. Often we encounter a somewhat restrained emotionality in the text that direct speech in the succeeding phrase as an emotional outcry of the assembly: «[…] corresponds to musical contrasts of medium intensity like the one just described. omnes clamauerunt dicentes». The shout «Benedictus deus!» however is given a rather unspectacular melodic segment at the very end of the respond. In examples of this Example 6 . R. Cum esset Petrus, Cao 6363. F-Pn lat. 12044, fol. 150v kind, we find a seeming contradiction between the emotional text and its setting, which R. [1] Cum esset Petrus in cruce, [2] venit ad eum turba multa, [3] maledicens cesarem; [4] et fecerunt appears as clear as the correspondence between emotion and music in examples planctum magnum ante crucem. 1 and 2. However, neither group is very representative. They range more or less at the periphery of what can be described as the ‘usual’ setting of direct speech in the responsories. In order to understand this ‘normal behaviour’, we first have to look more closely into the texts. When we talk about direct speech in the responsories from the perspective of meaning and emotion, we should generally be aware that by no means all of these texts have a distinctly ‘emotional’ character. Au contraire, many seem fairly calm in tone. This is especially true for most of God’s direct speech. In the responsories, God talks nearly always calmly and with majesty, he does not call or shout.14 The tone of many speeches of the martyrs and other biblical figures is likewise characterised by profound calm and the absence of any drama, even in the most turbulent of circumstances. This becomes particularly clear in example 6, a Cao- responsory about the passion of Saint Peter.

In this text, dramatic action is not located in the apostle’s direct speech, but in the introductory account, which is developed in the first half of the respond. During Peter’s In my final example (7, p. 276), I would like to demonstrate another kind of relationship crucifixion, the populace insults the emperor in its grief: «uenit ad eum turba multa, between music and text that appears to be quite characteristic of many pieces – and maledicens Cesarem; et fecerunt planctum magnum ante crucem». Peter, however, of our methodological problems in understanding them. The responsory from the exhibits the typical controlled sovereignty of the classical martyr. He exhorts the populace eleventh-century office in honour of Valeria of Limoges tells us about her execution at from the cross, using words that point not to panic, pain, and despair but to comfort and the command of the furious heathen duke Stephen. We encounter two introduced direct heavenly joy. See phrases 6 and 7: «*Nolite flere, sed gaudete mecum, quia uado uobis speech segments in this piece. The first, in phrase 4, represents Valeria’s prayer «Lord, parare locum». According to the theory of rhetoric, direct speech has the function of receive my spirit». Can we understand this prayer as emotional? We are inclined to do so, but we only have weak evidence for this. Phrase 3 informs us about Valeria’s gesture

14. An exception is the responsory Cao 6537: R. Dum deambularet Dominus in paradisum ad auram post meridiem, clamavit et dixit: Adam, ubi es? […]. 15. See Lausberg (1960, pp. 407-411, paragraphs 820-825).

274 275 CANTUS PLANUS 2014 HANKELN — THE ARTICULATION OF DIRECT SPEECH IN RESPONSORIES

Example 7. R. Cumque ducis. Valeria-historia. F-Pn lat. 903, fol. 201 In phrase 8 the final melisma over «gloria» comprises the same ambitus. I am doubtful, however, whether we can definitely characterise the direct speech in this chant as emotional in the ‘dramatic’ sense. If this evidence of the emotional quality is lacking, we have hardly any methodological basis upon which to determine an emotional motivation of their musical settings. What we can say in these and many other chants is, however, that the direct speech certainly constitutes the semantically most important passages in this responsory. In these cases I would thus prefer to talk about a melodic emphasis of words of central importance, rather than talk about ‘emotional’ music.

Conclusion Direct speech in its fullest form poses the compositional challenge of incorporating two text segments with a different content and emotional attitude, which are separated by a pause, into the traditional overall layout of the responsory. Direct speech thus belongs to those textual phenomena which – like enjambment, or questions – require specific formal musical reactions. The phenomenon of direct speech therefore offers the possibility of serving as an analytical marker for plainchant composers’ individual solutions to this textual problem. In this report about work in progress, I hope to have shown some of these solutions and their general background. Future research will perhaps be able to differentiate our knowledge about these solutions further.

Bibliography

Babb Warren - Palisca Claude V. (1978), Hucbald, Guido, and John on Music. Three medieval treatises, New Haven-London: Yale University Press. of prayer: the girl stretches her arms up to the sky, «virgo expansis in celum manibus Bailey Terence ed. (1979), Commemoratio brevis de tonis et psalmis modulandis, Ottawa: The University of Ottawa Press (Études médiévales de l’Université d’Ottawa, 4). oravit». This seems intense, certainly, but we cannot be sure. Praying with outstretched arms was the usual prayer gesture during the Middle Ages, at least until the thirteenth Barrasch Moshe (1987), Giotto and the language of gesture, Cambridge: Cambridge University Press. 16 century. In phrase 4 we read nothing about Valeria’s prayer having a dramatic tone: Dobszay László - Szendrei Janka eds. (2013), Responsories, 2 vols., Budapest: Balassi Kiadó. «oravit dicens» – to me this means at least that she is not crying or shouting. In phrase 6 a voice from heaven answers Valeria’s prayer: «ecce, o Valeria, mox uidebis Xpistum, Felder Hilarin (1901), Die liturgischen Reimofficien auf die Heiligen Franciscus und Antonius gedichtet und componiert von Fr. Julian von Speier (c. 1250), Freiburg, CH: Veith. sponsum tuum» (behold, O Valeria, soon you will see Christ, your bridegroom). The introductory unit in phrase 5 does not characterise the tone of this direct speech at all. We Fellerer Gustav ed. (1972), Geschichte der katholischen Kirchenmusik, 2 vols., 1, Kassel: Bärenreiter. are just told that a voice was heard from heaven: «de celis audita est». In both phrases 4 Hankeln Roman (2010), Schwerter und Pflugscharen: Zum Reflex des Geschichtlichen in der liturgischen and 6, the start of the direct speech is marked musically by a sharp contrast in register; Einstimmigkeit des Mittelalters in Rainer Kleinertz - Christoph Flamm - Wolf Frobenius eds., Musik the music of the direct speech initially lies a fifth higher than that of the preceding des Mittelalters und der Renaissance, Festschrift Klaus-Jürgen Sachs zum 80. Geburtstag, Hildesheim: Olms (Studien zur Geschichte der Musiktheorie 8), pp. 93-114. introductory phrases. The second direct speech marks, in addition, what is certainly the chant’s musical climax. The start of phrase 7 touches the highest note g and leads – ed. (1998), Historiae Sancti Dionysii Areopagitae. St. Emmeram, Regensburg, ca. 1050/16. Jh., Ottawa: Institute of Mediaeval Music (Musicological Studies 65/3). immediately down in a scalar movement to the low F, crossing the interval of a ninth. Hanssens Jeanne Michelle ed. (1948), Amalarii episcopi opera liturgica omnia, 3 vols., Tomus 1, Introductio. 16. See the chapter Prayer, in Barrasch (1987, pp. 56-95, here pp. 58-59). Opera minora, Città del Vaticano: Biblioteca apostolica vaticana (Studi e testi 138).

276 277 CANTUS PLANUS 2014

Danette Brink Helsen Katherine Eve (2008), The Great Responsories of the Divine Office: aspects of structure and transmission. Phil. Dissertation, Universität Regensburg. http://epub.uni-regensburg.de/10769/. Simeon: the long journey to Trier and Sainthood

Hesbert René Jean ed. (1970), Responsoria, versus, hymni et varia in Corpus Antiphonalium Officii, 6 vols., Roma: Herder, vol. 4 (Rerum ecclesiasticarum documenta: Series maior, Fontes 10).

Historiae (1995-), Zsuzsa Czagány - Barbara Haggh - Roman Hankeln eds., Lions Bay: The Institute of Mediæval Music (Musicological Studies 65/1-27).

Hughes Andrew (2011), The versified office: sources, poetry, and chants, 2 vols., Lions Bay: The Institute of Introduction Mediaeval Music (Musicological Studies 97/1-2).

Huseby Henning (2014), «Herskerens musikalske bilde». Om tekst og musikk i offisiet til St. Oswald av A saint is a deceased person who once excelled in virtue. A saint is one who possessed faith, hope Northumbria, Trondheim: Ntnu, Norwegian University of Science and Technology. and love, demonstrated wisdom and justice, exercised moderation and perseverance. A person who occasionally manifested these virtues can make no claim to sainthood, but only he who Jammers Ewald (1934), Das Karlsoffizium ‘Regali natus’, Strasbourg: Heitz; reprint 1984, Baden-Baden: persevered through his entire life, to a heroic degree, […] under difficult circumstances and with Koerner (Sammlung musikwissenschaftlicher Abhandlungen 14). a cheerful heart (Mulder-Bakker 2002, p. 3).

Johner Dominicus (19532), Wort und Ton im Choral, Leipzig: Breitkopf & Härtel. This modern definition of a saint, as described by Anneke Mulder-Bakker, includes Lausberg Heinrich (1960), Handbuch der literarischen Rhetorik. Eine Grundlegung der Literaturwissenschaft, München: Hueber. qualities such as virtue, faith, wisdom and justice that are universal, and were surely just as important during the Middle Ages when describing a holy person as they are Parkes Malcolm Backwith (1993), Pause and effect. An introduction to the history of punctuation in the West, now. The Archbishop of Trier, Poppo (1016–†1047), would have highlighted qualities Berkeley-Los Angeles: University of California Press. such as these in his appeal for canonization of the hermit, Simeon, to the Pope. By Schlager Karlheinz (2000), Ars cantandi - Ars componendi. Texte und Kommentare zum Vortrag und July 1, 1035, when Simeon died in the tower of the Porta Nigra and Archbishop Poppo zur Fügung des mittelalterlichen Chorals in Thomas Ertelt - Frieder Zaminer eds., Geschichte der Musiktheorie, 11 vols., Darmstadt: Wissenschaftliche Buchgesellschaft, vol. 4: Die Lehre vom started on his quest to have him canonized, Simeon had lived a long and sometimes einstimmigen liturgischen Gesang, pp. 217–293. exciting life, filled with events and miracles that would provide Poppo with useful fodder for constructing a vita, and certainly for fostering a cult. Simeon is one of the Slocum Kay Brainerd (2004), Liturgies in honour of Thomas Becket, Toronto: University of Toronto Press. ‘newer’ saints from the area around the Mosel River. Simeon’s feast day, June 1, first Snoj Jurij ed. (2003), Two Aquileian Poetic Offices, Ottawa: Institute of Mediaeval Music (Musicological appears in a Trier liturgical calendar, dating from the year 1128, from the monastery Studies 65/8). of Saint Simeon. One of the oldest sources containing a vita in honour of Simeon is Wagner Peter (1912), Einführung in die gregorianischen Melodien. Ein Handbuch der Choralwissenschaft, 3 D-TRs Ms. Nr. 118, a so-called Sammelbuch. Directly following the vita is a full office vols., Leipzig: Breitkopf & Härtel, vol. 2: Zweiter Teil. Neumenkunde; reprint 1962, Hildesheim: Olms. with adiastematic notation for this hermit-saint. This paper begins by presenting – (1921), Einführung in die gregorianischen Melodien. Ein Handbuch der Choralwissenschaft 3 vols., biographical background information on the saint, and how he came to be connected Leipzig: Breitkopf & Härtel, vol 3: Dritter Teil: Gregorianische Formenlehre; reprint 1962, Hildesheim: to the city of Trier. The office texts and their relationship to the saint’s vita will then be Olms. briefly discussed, before moving on to musical analysis of the chants composed in his – (1931), Die Gesänge der Jakobusliturgie zu Santiago de Compostela: aus dem sog. Codex Calixtinus, honour as found in the oldest known source with diastematic notation for this office, Freiburg, CH: Universitäts-Buchhandlung (Collectanea Friburgensia. Neue Folge 20). the antiphoner D-TRb Ms. Nr. 480.

Vita Sancti Symeoni Simeon was born in Syracuse, Sicily, to Christian parents, in ca. 990. As a seven-year- old, he was sent to Constantinople to further his education; some historians argue that he might have gone there to study law (Heikkilä 2002, p. 116). He later moved to Jerusalem, where he acted as a guide to pilgrims in the Holy Land. Not content with this position, and probably in some respects due to his childhood dream of serving God as a hermit, he worked as a servant for a hermit in Jordan, before entering the

278 279 CANTUS PLANUS 2014 BRINK — SIMEON: THE LONG JOURNEY TO TRIER AND SAINTHOOD monastery of Saint Maria in Bethlehem, for two years, where he was ordained as a city became his new home. There he met the Archbishop of Trier, Poppo, who asked deacon. With the monastery abbot’s permission, he moved to a monastery at Mount Simeon to accompany him and act as his guide on a pilgrimage to Jerusalem. Simeon Sinai, Saint Catherine, where he spent two years at the monastery and two years agreed, and set out on his last, great voyage, accompanying the Archbishop on his as a hermit on the shores of the Red Sea. He then moved back to the monastery on journey (aa ss, p. 89). Upon their return to Trier, Poppo granted Simeon the choice of Mount Sinai (Acta Sanctorum, June 1, henceforth aa ss, pp. 86-87). Richard ii, Duke any place in his archdiocese to make his new home. Simeon had another vision which of Normandy, had promised funds to the monastery at Mount Sinai (aa ss, p. 88)1 led him to choose the Porta Nigra. Therefore, with the Archbishop’s blessing, he was and the abbot of Saint Catherine decided to send Simeon to collect it.2 From the start, walled within a room in the tower on the feast day of Saint Andrew, finally achieving things did not go well with Simeon’s voyage. While travelling on the Nile, his boat was his youthful desire to become a hermit. After his retreat into the Porta Nigra, Simeon attacked by pirates, forcing him to change his itinerary and travel to Antioch instead.3 could receive and have discussions with visitors, so he was not completely cut off from There he met a group of western Europeans from Aquitaine, Lorraine, the Rhineland, the outside world. He received visits from Eberwin and from his servant, Stephanus. Flanders, and Normandy, who were travelling on a pilgrimage to Jerusalem together, However, his time in the Porta Nigra was far from peaceful. Simeon was plagued by and Richard and Eberwin, abbots from Verdun and Trier, in Antioch. Simeon decided demons and an apparition of a wild animal; a great flood ravaged the city of Trier, and to wait in Antioch for the pilgrims’ return from Jerusalem in order to accompany them the people accused Simeon of harbouring demons and pelted his cell with stones. His on their homeward journey. This journey probably took place in around March or only refuge was in prayer (aa ss, p. 90). April, 1027. While in Antioch, Simeon met another monk called Cosmas, who decided to join him as a travelling companion (aa ss, p. 89). Simeon foresaw his own death and gave a verbal testament that he expressly wished not to be buried in the Trier cathedral or a monastery, but that he wanted to be Returning westward, the group was stopped in Belgrade, and Simeon and Cosmas entombed in the Porta Nigra. Eight days after he had announced that he would die were separated from the pilgrim group and forbidden to continue on their way.4 A and after much suffering, he breathed his last. Simeon expired in the Porta Nigra on possible alternative route to reach Francia, was to cross the Adriatic Sea and travel by the first day of June, 1035 aa( ss, p. 90). Archbishop Poppo wrote to Pope Benedict ix way of Rome. Whether a stopover took place in Rome or not, Heikkilä suggests that shortly after Simeon’s death, to appeal for Simeon’s canonization. The preparation the two travelling monks arrived in Angoulême in May 1027. They were hosted by of an altar at Simeon’s gravesite in as early as November, 1035, is an indication of Count William, whom Simeon had met in Antioch. The Count also introduced them the archbishop’s confidence that the pope would grant his request. To help argue to Ademar of Chabannes during their stay. Unfortunately, Cosmas died in Angoulême, his cause, Poppo instructed Eberwin to write a Vita about Simeon’s life and death, and Simeon had to travel on alone to Rouen (aa ss, p. 89; Heikkilä 2002, pp. 118-119). including the miracles performed before and after his death.5 Simeon was canonized Symeon arrived in Rouen around July or August 1027. There he was met with more by the Pope no later than 1039, although the exact date is unknown.6 The haste with bad news: Richard ii, who had promised money to the monastery, was dead, and his which Poppo initiated the appeal for the hermit’s canonization has a certain sense son, Richard iii, did not intend to honour his father’s pledge. Before setting out from of urgency about it. In the eleventh century, Trier’s spiritual status was so well- Egypt, Simeon had had a vision that this particular voyage would be long and painful, established that it was known as a civitas sancta. This, however, did not keep the and in the end bear no results. Despite having shared this presentiment with the abbot bishops of other dioceses, such as Cologne, Mainz and Magdeburg, from competing of Saint Catherine, the abbot still ordered him to go (Krönert 2003, p. 470). From Rouen, with the city of Trier to become the undisputed leader of the Church in the region. Simeon probably travelled to Verdun to visit his friends, the abbot of the monastery Moreover, in the tenth and eleventh centuries, many of the Archbishops of Trier of Saint-Vanne, Richard, and Eberwin, who was (probably) at that time the abbot of had very close ties not only with the Pope, but also with the Emperor. However, monasteries in Tholey, Trier and Verdun. He stayed with Eberwin in Trier, and the even this failed to deter secular leaders, such as the Dukes of Luxemburg, from waging land wars against the Archbishop of Trier, especially in the time of Poppo 1. Although it is known that Richard ii gave grants to monastic houses under his ducal control, it is unclear why he would and his successors (Heikkilä 2002, pp. 73-74). Poppo’s predecessor, Megingaud, donate money to a monastic house in Egypt. was in conflict with Adalberto, a relation of the Dukes of Luxemburg and provost 2. Another unanswered question arises from this mission: why would the abbot decide to send a mere monk to collect this donation, in all probability, an important contribution to the monastery’s coffers? Perhaps Simeon’s Wanderlust and of the monastery of Saint Paulin, who named himself as Archbishop of Trier, while knowledge of foreign lands, gained through previous travels, might have persuaded the abbot to send this particular monk.

3. The modern Antakya, Turkey. For the purpose of this paper, the medieval name, Antioch is kept.

4. Heikkilä (2002, p. 118) suggests that they were stopped in Belgrade, because they were thought to be spies of the 5. aa ss, chapter 4: «Miracula varia, meritis S. Symeonis patrata, post mortem et in vita». See also Krönert (2003 p. 471). Emperor of Byzantium. Krönert (2003, p. 471) suggests that it was because of his Greek origins that Simeon was not allowed to continue the journey. Of course, in Heikkilä’s theory, the reason for the fact that Simeon was thought to be a spy for the 6. According to Heyen (2002, p. 261), the Pope canonized Simeon during the Christmas season 1035, and Poppo received Byzantinian government could be related to his Greek origins. the news in early 1036. See also Heikkilä (2002, p. 170).

280 281 CANTUS PLANUS 2014 BRINK — SIMEON: THE LONG JOURNEY TO TRIER AND SAINTHOOD

Megingaud enjoyed the support of Emperor Henry ii. This brought the two factions responsory, Gaude plebs Treverica, the Benedictus antiphon, Laus honor et gloria, and into bloody conflict, until Megingaud eventually established his administrative the second vespers Magnificat antiphon, Sancte Symeon nobis, are all laudatory in centre in Coblenz. It was Poppo who finally managed to subdue Adalberto. nature. Although the antiphons Laus honor et gloria and Sancte Symeon nobis refer to miracles, they are more general in nature, and the Magnificat antiphon is a summary Given this political clime, it is perhaps no wonder that Poppo wanted to cement both of the miracles used by the author for the lauds antiphon texts. Responsories seven his position and that of his city as archbishopric, through the canonization of saints and eight, Mox Symeon factus and Mors beati Symeonis, provided a brief comment on and the cultivation of their cults. In fact, the cult of Simeon was only one of many to Simeon’s death. be established at this time. Others included the cults of Celsus, Cuno and the martyrs of the monastery of Saint Paulin, as well as the cult of the Apostle Matthias a few years There is a discrepancy between the chant texts and the Vita texts. In the seventh later. A double church and monastery was founded and built around the Porta Nigra.7 antiphon of matins, the author describes Simeon’s last great journey, in which he From the beginning, the establishment was referred to as ecclesia sancti Simeonis. In accompanied Archbishop Poppo to Jerusalem. According to Eberwin, Simeon travelled early 1036, a small college for priests (Priesterkollegium) was established by Poppo all the way to Jerusalem with Poppo, returning with him later to Trier. In the chant to promote the devotion of the cult of Simone as well as to support and assist the texts, however, the writer narrates that Simeon accompanied Poppo to Antioch, but he pilgrims that flocked to worship at his grave. Only in 1041 does the name ‘Saint Simeon decided not to proceed all the way to Jerusalem, but to wait for Poppo’s return there Monastery’ appear; the charters for this institution can be dated back to 1048 (Heyen (table 1). A possible explanation for the writer of the texts having mixed up Simeon’s 2002, pp. 262 ff.). different voyages is that they were written long after the fact, when memories had faded, and none of Simeon’s contemporaries had survived to rectify the error. Since The chant texts: literary form and relationship to the Vita Eberwin would not have easily fallen prey to such confusion, it is possible that he is According to Heikkilä (2002, pp.148-168), the Vita and Miracula can be found in around not the author of the texts of the office for Saint Simeon. fifty sources. The Vita is divided into three main parts. Part i, which is dedicated to Simeon’s activities and accomplishments during his lifetime, is further divided into Musical analysis three subparts: his childhood, the various journeys he undertook during his lifetime, Simeon died in 1035, and it is reasonable to expect that the office in his honour was and the time he spent as a hermit within the Porta Nigra. Part ii describes his death, and written shortly after his death. Part iii is about the miracles he performed, both before and after his death. Eberwin 8 is acknowledged as the author of the Vita Symeonis. The most interesting source is Table 1. Comparison between antiphons vii and viii and the Vita Symeonis, chapter 3 D-TRs Ms. Nr. 118. The provenance of this manuscript is Trier, although it is not clear from which monastery. A stamp on the inside of the binding shows that it belonged Vita texts Chant position Chant texts to the monastery of Saint Eucharius during the seventeenth and eighteenth centuries. Chapter 3 Interea Dominus Poppo, Antiphona Ceptum iter Symeon Although the manuscript contains different parts dating from the ninth to the fifteenth (15), 89 Archiepiscopus, sanctae vii & viii impleverat sed mox cum century, Heikkilä dates the part containing the life and miracles of Simeon to the Trevirensis Ecclesiae Popone Treverorum presule eleventh century (Heikkilä 2002, p. 151).9 The manuscript contains not only the Vita strenuus provisor, Ierosolimam tendente Symeonis by Eberwin, but also the office for Saint Simeon with adiastematic notation. audiens devote Prophetam remeare ceperat. de Christo dicentem, Keuffer, in his catalogue of the manuscripts in the Trier Stadtbibliothek, mistakenly Et erit sepulcrum Qui cum simul Anthiochiam identifies the musical entries as «Oration und Sequenz zu Ehren des h. Simeon, […]» ejus gloriosum, gratia venirent divina prefiguratione (Keuffer 1891, pp. 8-9). The chant texts are written in prose and are based on the Vita, orationis Hierosolymam Symeon restitit episcopus ivit: huncque famulum Ierosolimam it et rediit. although not repeated verbatim. The texts of the First Vespers Magnificat antiphon, Dei eundo et redeundo Venerantes et dignam, the invitatory antiphon, Iubilemus regemque regum, the ninth secum comitatorem ac conviatorem habuit: eique post reditum, in suo

7. «Über die Gründung des Stiftes St. Simeon gibt es keine formale Urkunde; es ist auch nicht anzunehmen, daß darüber Episcopatu quocumque überhaupt eine Urkunde ausgefertigt worden war» (Heyen 2002, p. 261). vellet manendi facultatem obtulit, et libentissime 8. «Auctore Eberwino, Abbate S. Martini Treviris» (aa ss, p. 86). concessit. 9. Keuffer (1891, p. 9) confirms this: «[…] von einer Hd. des 11. Jhrd. […]».

282 283 CANTUS PLANUS 2014 BRINK — SIMEON: THE LONG JOURNEY TO TRIER AND SAINTHOOD

Through comparison of the notation of the two sources, it was found that the neumes Table 2. Modal order of antiphons of D-TRs Ms. Nr. 118 correspond, for the most part, to the melodies of D-TRb Ms. Nr. 480. Even when the melody is sometimes different per syllable, for example, using Position Incipit Mode 10 11 a pes instead of a clivis in D-TRs Ms. Nr. 118, or one note instead of three, the 1V-Am Venerantes et dignam 4 general shape of the melody is retained. The writer of the office in D-TRs Ms. Nr. M-I Jubilemus regem que 2 118 does not keep to any set ideas concerning the use of the virga and the punctum. M-A1 Iste Symeon quem 1 Mostly the punctum would be used to indicate a lower tone, but that is not always the case, for example in the third responsory, Quadam nocte demones, the second phrase M-A2 Qui patrem suum 2 beatum: the melody in D-TRb Ms. Nr. 480 moves from D to E and the E is indicated M-A3 Postquam juvenilem aetatem 3 by a punctum. This of course could also mean that the melody in the D-TRs Ms. Nr. M-A4 Vir Dei ut 4 118 melodic version goes to C, instead of E. Another example is the fifth antiphon, M-A5 Mane facto magistrum 2 Mane facto magistrum. In the second phrase, the melody moves upwards from D, E, F and G (et que in nocte), notated in D-TRs Ms. Nr. 118 as punctum, virga, punctum, M-A6 Cujus consilio magis 7 virga. Several musicologists, chiefly László Dobszay, David Hiley and Roman Hankeln, M-A7 Ceptum iter Symeon 4 have identified a number of musical and textual characteristics that can be used to M-A8 Qui cum simul 1 distinguish later office chants (the so-called ‘post’ or ‘neo’-Gregorian chant) from the M-A9 A Deo Symeon praemonitus 6 so-called ‘traditional’ Gregorian chant repertory. The ‘traditional’ repertory shows L-A1 Cum jam beati 1 a tendency towards the use of the tetrachordal system and melodic building-blocks for composing the melody. The ‘neo’-Gregorian repertory tends to contain numerous L-A2 Mox ipsius elemosinaria 3 late stylistic characteristics together in the same chant, i.e., the use of goal pitch- L-A3 Que coram multitudine 4 oriented melodic phrases, the use of the so-called Gallican ending, scale movement L-A4 Quidam puer diu 5 (or segments), the use of note groups over one syllable and frequent interval leaps L-A5 Cujusdam filia a 6 of over an interval of a fifth or more. The absence of traditional verse tones with the L-Ab Laus honor et gloria 7 responsories is also indicative of a new composition. Text can also help determine the era in which a cycle of chants may have been composed. Prose was the form of choice 2V-Am Sancte Symeon nobis 1 for chant texts written in the traditional Gregorian style, and rhymed verse in chant texts from this period is rare, although it does appear. During the eleventh century, rhymed prose became more popular, and by the end of the twelfth century, accentual Table 3. Modal order of responsories rhymed verse had come into vogue. All of these characteristics will not, of course, be present in all chants, making them appear homogeneous. Some chants may display Position Incipit Mode more ‘modern’ stylistic characteristics, whilst others may have a mixture of traditional M-R1 Beatus Symeon compertis 7 and newer features. These can then help to determine where to place these cycles of M-R2 Dum ad huc istum laborem 2 offices chants on a hypothetical composition timeline, assigning the chants to different M-R3 Quadam nocte demones 1 ‘layers’ according to their stylistic approach. M-R4 In heremo beatus 4

The modal order of the antiphons in the office for Saint Simeon is not in sequence, M-R5 Post multum temporis 6t* except for the antiphons in the first nocturn (first-third modes); and the antiphons in M-R6 Deinde tendens 8 the lauds, from the second antiphon through to the Benedictus antiphon (third-seventh M-R7 Mox Symeon factus 2 modes, table 2). The responsories, too, are not in modal order (table 3). M-R8 Mors beati Symeonis 7

10. For example, in the Benedictus antiphon Laus honor et gloria, in the fourth phrase, laudibus, the second syllable in M-R9 Gaude plebs Treverica 2 D-TRs Ms. Nr. 118 is a pes compared to a clivis in D-TRb Ms. Nr. 480. 11. For example, the first lauds antiphon, Cum jam beati, fourth phrase sentire. *t = transposed

284 285 CANTUS PLANUS 2014 BRINK — SIMEON: THE LONG JOURNEY TO TRIER AND SAINTHOOD

Melodies of eighteen chants use a full octave range (although eight of these do not Example 2. Responsory Beatus Symeon compertis correspond with the range of the melodies’ respective modes, usually occurring with chants in the third and fourth modes), while the ambitus of five chants are wider than an octave. The use of goal pitches in each chant is inconsistent. Of the thirty-six chants in this office, only eight chants have goal pitches on three quarters or more of their polysyllabic words. The use of goal pitches on polysyllabic words are generally less frequent in the antiphons than in the responsories, exceptions being the second matins antiphon, Qui patrem suum, the fourth lauds antiphon, Quidam puer diu, and the Benedictus antiphon, Laus honor et gloria. Due to limits of space, only two examples will be analysed here. In the Benedictus antiphon, Laus honor et gloria, (example 1), the melody is not formed through the use of repetitive melodic fragments, but rather through melodic movement centred around the final G and the dominant d. Although the melody does move in the central scale segment of the mode, it is rather more focused on the upper modal framework of the mode; many of the melodic high points focus on the high D, and the melody touches numerous times on f and three times on g. Such frequent use of the tone f may seem strange. However, in my opinion, the f is not so much the melodic high point, as is the D to which it almost always leads, although sometimes through an intermediate (or passing) tone, for example, on gloria and eterna. Most of the words end either on the final (G) or on the fifth above it (d). Exceptions are laus, tibi, qui, tot and et.

Example 1. Antiphon Laus honor et gloria

Large scalar fragments to create movement are not often used in this chant. Where these fragments do occur, it is usually to facilitate movement between ‘important’ modal tones, such as the final and the fifth above the final, for example Domine, where the melody moves from the high g to the d through a scalar motion, moving from the octave above the final to the fifth of the mode. Another example is secula, where the melody moves from the fifth above the final to the final through an interrupted stepwise motion and the use of a so-called Gallican cadence. The melody moves through the whole range of the seventh mode, up to the high g, and it never moves lower than the F, which is not expected from a melody in the seventh mode. The movement of each phrase also rests either between the final (G) and the fifth Although not frequent, the tonal jumps of a fourth and a fifth (Laus honor and sanctum above it (d) or between the fifth and the final (g) above that tone. The melody of the respectively) are representative of a later compositional style. The so-called Gallican first responsory, Beatus Symeon compertis (example 2), is not made up of motivic cadence features often (honor, gloria, secula, dignatus es, miraculis, honorari, laudibus). elements, but rather moves in ‘scalic’ motion between the tonic G and the dominant

286 287 CANTUS PLANUS 2014 BRINK — SIMEON: THE LONG JOURNEY TO TRIER AND SAINTHOOD d. Most of the melodic ‘figures’ that do feature are built either around the tonic G or The office for Saint Simeon is a good example of how sainthood was used for political the dominant d. For example, in the Beatus, the melodic motion (and figure) is built purposes. Archbishop Poppo had a goal of re-establishing the ecclesiastical prominence around the fifth above the final, d. of Trier, so he used Simeon’s death (and by extent his life) to achieve it. At the time of Simeon’s death, being a hermit had become a popular way for Christians to become Most multisyllabic words end on either the final or the fifth (exceptions being strictionis, closer to God. If Poppo could capitalize on Simeon’s death, having him canonized soon which ends on F and sanctitatis, which ends on c). Most one- or two-syllable words afterwards, when the miracles that occurred at his cell were still fresh in the minds do not ‘end’ on the final or the fifth, and consist of only one neume (exception:ad of the people, he would be sure to have a steady flow of pilgrims coming to his newly which consists of a group of neumes starting and ending on the fifth, d, and se ending built church in honour of the saint. The composition of an office for Saint Simeon on d). Not all words start on the fifth or final. However, Stäblein’s theory of «modal would surely have played a role in achieving this goal. The saint’s life and deeds would integrity in individual words» can be applied. Stäblein states that accented syllables have been recalled on his feast day through the singing of the chants composed in would be «auskomponiert» – given more importance – and unaccented syllables less his honour, reminding those who were singing them to try and live up to the saintly so (Stäblein 1975, p. 162). Estelle Joubert (2009, p. 44) sums up Stäblein’s theory as values expressed in their prose. « […] individual words in plainchant for offices begin and end on pitches that are modally important, thereby creating stability and overall modal integrity in the item». The responsory verse tone has echoes of the original seventh mode verse tone, as the melody centres around the fifth above the final, d. The differences that can be mentioned between the original verse tone and the verse tone used in this office is the tonal space the melody occupies. The original seventh mode verse tone moves mainly between G and d, occasionally moving beyond to e and f. The verse tone used in Beatus Symeon compertis often moves to the high f, even using the high g as the ‘high point’ in the melody. The ending also greatly differs from the original verse tone: the original verse ends on h and the Simeon verse tone moves downwards from d in Bibliography a ‘stepwise motion to the F below the final before resting on the final, G. None of the Vita Symeonis (1876), in Godefroid Henschen - Daniel Paperbroch eds., Acta sanctorum […] Junii tomus other responsories make use of the traditional verse tones either. primus continens sex primos dies […], editio novissima, Paris-Rome: Victor Palmé, pp. 86-101.

Heikkilä Tuomas (2002), Vita S. Symeonis Treverensis: Ein hochmittelalterlicher Heiligenkult im Kontext, Conclusion Helsinki: Tiedekirja. The office for Saint Simeon in D-TRs Ms. Nr. 118, and its continued use, as reflected Heyen Franz-Josef (2002), ‘Das Stift St. Simeon in Trier’, Germania sacra, neue Folge 41: Die Bistümer der in its inclusion in the antiphoner D-TRb Ms. Nr. 480, which was used at the Trier Kirchenprovinz Trier: Das Erzbistum Trier 9. cathedral, shows that the liturgical veneration of Simeon can be traced back to the Joubert Estelle (2009), ‘New music in the Office of Thomas Becket from the Diocese of Trier’, Plainsong and eleventh century. The music for this office does not show an abundance of new stylistic Medieval Music 18/1, pp. 33-60. characteristics. The chants are not in modal order, although most of their melodies range over an octave. Goal pitches play an important role in the melodic movement, Keuffer Max (1891), Die Kirchenväter-Handschriften der Stadtbibliothek zu Trier. No. 113 bis 214 des Handschriftenkatalogs, Trier: Lintz (Beschreibendes Verzeichnis der Handschriften der Stadtbibliothek rather than tonal building-blocks. There are, however, few instances of big interval zu Trier 2). leaps, or rapid movement through scales or tone fragments, especially in the case of Krönert Klaus (2003), La construction du passé de la cité de Trèves: VIIIe - XIe siècles. Etude d’un corpus the antiphons. The text of the office is also in prose. The office for Saint Simeon might hagiographique, Dissertation, Université Paris X - Nanterre. be classified as part of the middle layer of the ‘neo’-Gregorian chant style, where the chant melodies have moved away from the old ‘conventional’ melodic system, but have Miesges Peter (1914), Der Trierer Festkalender. Seine Entwicklung und seine Verwendung zu Urkunden­ datierungen. Ein Beitrag zur Heortologie und Chronologie des Mittelalters, Dissertation, Rheinische yet to loosen ties with the old tradition completely. It is not plausible that Eberwin was Friedrich-Wilhelms-Universität, Bonn. the author of the chant texts in honour of the hermit Simeon. It is, however, possible Mulder-Bakker Anneke B. (2002), The Invention of Saintliness: Texts and Contexts in Anneke B. Mulder- that the office chants and texts were composed in the latter part of the eleventh century, Bakker ed., The Invention of Saintliness, London-New York: Routledge, pp. 3-23. a theory which the textual and musical analyses support. Stäblein Bruno (1975), Schriftbild der Einstimmigen Musik in Werner Bachmann ed., Musik des Mittelalters und der Renaissance, 9 vols., Leipzig: Deutscher Verlag für Musik, vol. 3 (Musikgeschichte in Bildern 3).

288 289 CANTUS PLANUS 2014

VII Regional and local traditions

290 291 Raquel Rojo Carrillo

The vespertini as witnesses to the transmission of Old Hispanic chant in traditions A and B

Ever since the mid-twentieth century, scholars have classified the Old Hispanic rite manuscripts into two liturgical traditions. The names most commonly used to refer to these groups, tradition A and tradition B, were proposed by Jordi Pinell (1965). Traditions A and B share some of their material. However, certain differences in the liturgical assignments, melodies and/or texts of this shared material, as well as the fact that some chants, readings and prayers are unique to each tradition, prompted this classification. The few existing theories concerning the origin and transmission of Old Hispanic chant rely on the distinction drawn between traditions A and B (e.g., Randel 1969a, pp. 93-100; Pinell 1972, 1978, and 1998; Janini 1977; Hornby and Maloy 2003, pp. 303-314).

Most of the Old Hispanic manuscripts with musical notation belong to tradition A.1 They comprise almost forty manuscripts copied from the tenth century to the early thirteenth century in different parts of the Iberian Peninsula (modern Spain and Portugal), and were produced for different types of institutions, including monasteries, cathedrals and parishes. By contrast, just three notated manuscripts preserve tradition B: the Lenten libri mistici E-Mn MSS/10110 and E-Tc 35.5, and the fragment of liber misticus for Post-Pentecost time E-Tm n.i. 1325.2 E-Mn MSS/10110 has a colophon (fol. 120v) specifying that it was copied for the parish of Santas Justa y Rufina of Toledo, located in central Spain. The two other tradition B manuscripts have been ascribed to this parish by some authors (e.g., Pinell 1965 and 1998), despite the fact that medieval evidence proving these ascriptions has not surfaced, as shown in Rojo Carrillo (2018).3 The tradition B manuscripts with musical notation are believed to have been copied between the mid-thirteenth century and the beginning of the fourteenth century (I adopt the dates given by Mundó, 1965, for the Toledo-related manuscripts belonging to both traditions). The extant B manuscripts

1. An up-to-date and comprehensive catalogue of Old Hispanic manuscripts which specifies, inter alia, their liturgical tradition, as well as their confirmed or ascribed copy-dates and provenances, is available at the Spanish Early Music Manuscripts Database (I authored and maintain this catalogue).

2. The two other fragments considered as belonging to tradition B (Pinell 1965, pp. 118 and 122), E-Tm n.i. 1326 (fragment of liber commicus) and n.i. 3.2 (fragment of liber hymnorum) lack musical notation. Brief explanations of the typology of Old Hispanic chant books (e.g. misticus, commicus, etc.) are included in each manuscript’s description at the Spanish Early Music Manuscripts Database .

3. Scholars have associated E-Tc 35.5 with the parish of Santas Justa y Rufina because its repertoire is very similar to that in E-Mn MSS/10110. They have proposed this provenance for E-Tm n.i. 1325 because this manuscript has a modern seal of this parish in several of its folios. However, authors such as Janini (1977) have noted that before arriving in this parish, E-Tm n.i. 1325 was preserved in other Toledan parishes. For more details see Rojo Carrillo (2018).

293 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT transmit the public (cathedral ordo) temporale liturgy for one complete liturgical psalmi, have shown that A and B relate differently in each genre. For example, some season only, Lent, and for just three days of another season, Eastertide;4 they also genres, such as the threni, are very similar in their A and B presentations on all transmit sections of the public liturgy of four sanctorale days falling in the Post- three levels, while other genres, such as the responsorial tones, present significant Pentecost time.5 In tradition A, the liturgy for these days (Eastertide temporale and differences and even subgroups within each tradition. Additionally, some genres Post-Pentecost sanctorale) have survived in more than one manuscript,6 but Lent have a subtler and more complex relationship between their A and B chants, is completely preserved in just one manuscript, the Antiphonary of León (E-L Ms presenting different degrees of similarity at each level; this happens with the Lenten 8). This manuscript originated in a very different context from that in which the psalmi, for example. Consequently, the results found in one genre are normally not B sources were copied: it was in all likelihood produced ca. 960 for a powerful transferrable to others. However, they provide useful information for determining cathedral – one linked with the royalty – of northern Spain.7 A comparison of a whether or not the twentieth-century classification of the Old Hispanic manuscripts selection of chants from the two liturgical traditions can help to determine whether into two traditions is actually applicable to the whole musical repertoire transmitted these groups of manuscripts were really distinct, or if, instead, they constitute an by these manuscripts. example of how a flexible liturgy could be adapted to two very different contexts. In what follows, I will identify the main levels at which chants shared by both The examined genre traditions can differ. After this, I will undertake a detailed comparison of two chants The vespertinus was the chant that opened the Old Hispanic vespers service. Vespers belonging to the vespertinus genre and shared by both traditions. By building on this were normally observed every evening as the first service of each liturgical day. comparison I will propose the possible causes underlying their differences. The vespertinus was, therefore, the very first chant of each office. In this position it introduced an important topic of the liturgical day. The vespertinus was also Different levels of relationship used to open second vespers, the last service on certain days. At second vespers, A and B chants of a same genre can relate differently at three levels: they can vary the vespertinus would remind the congregation of an important topic of the day in their liturgical assignments, in their texts, and/or in their melodies. Even within and/or have a praising role. For example, the vespertinus Inlumina domine vultum a single source, for example, we can find chants with identical texts but different for the first vespers of Palm Sunday is about learning through the contemplation of melodies and liturgical assignments, as well as chants with different texts but God, thus introducing the theme of initiation, which is appropriate for this day when similar melodies and liturgical assignments. For example, in E-L Ms 8 there are future Christians would be exorcised before their approaching baptism;8 and the three vespertini with the same text in their main section: Dominus inluminatio mea vespertinus Usque ad vesperum, assigned to second vespers of certain Sundays of et salus mea quem timebo. These vespertini, however present a different melody and Lent, including Palm Sunday, is a praising chant. liturgical assignment in each case: fol. 50, for Saint Eulalia; fol. 119, for Mondays of the first half of Lent; and fol. 13v for Mediante Sunday. This manuscript also Transcription of text Translation contains three vespertini that have different texts and similar assignments (Lenten Inlumina domine vultum tuum super serbos tuos Make your face shine upon your servants, weekdays), but that use the same melody: Principes persecuti, fol. 134v; Eripe me, et doce nos iustificationes tuas (Ps 118:135) O Lord, and teach us your laws. fol. 138v; and Conserva me, fol. 140. Moreover, results from previous genre studies VR Declaratio sermonum tuorum inluminet nos VR Let the declaration of your words enlighten us et intellectum dat parvulis (Ps 118:130) and give understanding to the children, by Randel (1969a) on the responsorial tones (i.e. melody-types used to chant the et doce nos iustificationes tuas (Ps 118:135) and teach us your laws. responsorial verses), and by Hornby and Maloy (2013) on the threni and Lenten Usque ad vesperum quam magnificata sunt Until the evening, how magnificent are opera tua domine (Ps 103:23-24) your works, O Lord 4. The Eastertide offices preserved in tradition B manuscripts, specifically in E-Tc 35.5, are: Easter Sunday, and the first Monday and Tuesday of Eastertide (i.e. the octave Monday and Tuesday; the latter day is incomplete).

5. Preserved in E-Tm n.i. 1325 only (a fragment which lacks liturgy for Lent): Saint John the Baptist, Saints Peter and Paul, Traditions A and B share certain features in the liturgical assignments for their Saints Simon and Jude, and Saint Martin. The only vespertinus assigned to sanctorale offices by tradition B manuscripts Lenten and Eastertide vespertini: (specifically to Saints Peter and Paul, and Saints Simon and Judas in E-Tm n.i. 1325) is preserved in incipit format without notation. Consequently, this paper builds on the extant notated repertoire given in B manuscripts, which is for Lent and the start of Eastertide, and on the repertoire for the same days in tradition A manuscripts. Both traditions preserve vespertini in two formats. The first consists of a single 6. Some of the Old Hispanic chant manuscripts with the Post-Pentecost time sanctorale are: GB-Lbl Add MS 30845, E-Tc 35.6 and E-SI Ms 5. section, apparently to be performed without repetitions (e.g., Usque ad vesperum,

7. The elegant textual and musical script, as well as the quality of the miniatures in E-L Ms 8 reveal that it was intended for a renowned recipient. Royal monograms in several of its folios show that it was later owned by the Leonese monarchy. Comprehensive descriptions of this manuscripts are included in Antiphonarium mozarabicum (1928); Brou (1954); and in 8. I am grateful to Dr Kati Ihnat for informing me that exorcisms were practised during the Old Hispanic Palm Sunday the Spanish Early Music Manuscripts database . office, as instructed in fol. 152v of E-L Ms 8.

294 295 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

above); the other contains a main section (equivalent to a respond) plus one or that the vespertini with verses are assigned only to Sundays, Holy Week and Eastertide, more verses, each of which is normally followed by a repetition cue that calls for a and that the one-section vespertini are assigned to second vespers of Sundays and responsorial performance (e.g., Inlumina domine vultum, above). weekdays of Lent. If we have a closer look at the texts of the vespertini for Lenten weekdays in E-L Ms 8 (highlighted in colour), we can see that the set for the first half During Lent, these two types of vespertinus are assigned according to the liturgical of Lent contains chants of praising character that employ the metaphors God/light and importance of the day and the service, as will be demonstrated below. However, sin/darkness, often referring to light-shift times of the day. The weekday vespertini the two traditions differ in the specific chants for each day. for the second half of Lent in E-L Ms 8, by contrast, evoke the theme of persecution and have a pleading character, normally using imperative verbs. Thus, through its At Eastertide, both traditions assign the same long vespertinus to Sundays and weekday vespertini, tradition A’s E-L Ms 8 follows the division of Lent into two halves, Easter octave weekdays, but the layout of this chant is different in each manuscript. emphasising the particular character of each half.11 It can also be seen that both sets In both traditions the organisation of Lenten weekday vespertini follows the of weekday vespertini follow the numerical order of the Psalter, although not always order of the Psalter (though somewhat loosely), but each tradition has its own strictly. This layout links certain days through the use of the same psalm number, and organisational pattern. even verse, in their vespertini: Mondays, Wednesdays and Thursdays in the first half of Lent, and just Wednesdays and Thursdays in the second half. Finally, the vespertinus The A and B Lenten and Eastertide temporale vespertini were crafted using psalmic for Easter and the beginning of Eastertide, Laudaberunt te domine, employs texts verses, following the same techniques, and sharing the same general goals. The praising the deeds of the Lord. In E-L Ms 8, the first appearance of this chant (fol. 179v) psalmic verses of these chants are rarely quoted exactly. Instead, the vespertinus is very likely both a vespertinus with five verses to be chanted on second vespers of sections normally consist of only part of a psalmic verse, or of a mixture of two Easter Sunday and, as noted by Randel (1969b, pp. 105-106), a list of five vespertini to fragments of psalmic verses, not necessarily following the same order of the Psalter. be chanted on other Eastertide offices. This is suggested by several pieces of evidence, The psalmic texts were adapted very likely to convey a topic appropriate for the including: a) the instructions for the observance Eastertide in E-L Ms 8 (fol. 179); b) liturgical day. For example, the initiation and persecution themes are typical of Lent the layout of Eastertide in this manuscript; c) the comparison of this chant with other in this and other genres, e.g. the threni, Lenten psalmi and responsories studied vespertini assigned to solemnities in this manuscript (all of them with more than one by Hornby and Maloy (2013). Despite these modifications, the vespertini that are verse); and d) the comparison of this chant with its cognates in other manuscripts.12 shared by both traditions present the exact same texts in A and B. However, only tradition A contains two sets of weekday vespertini with different thematic focuses Table 2 (p. 306) lists the totality of temporale vespertini preserved in tradition B corresponding to each half of Lent. manuscripts. As mentioned above, these sources coincide with E-L Ms 8 in giving vespertini with verses for Sundays (first vespers) and Eastertide, and one-section Finally, all but one tradition B vespertini also exist in tradition A manuscripts, vespertini for weekdays and second vespers. Tradition B manuscripts, however, only while several unica vespertini exist in tradition A’s Lent. It is worth noting here give one set of weekday vespertini for all weeks of Lent. Thus, in this season, the that at other times of the year, tradition A’s E-L Ms 8 also has vespertini which do same vespertinus would be sung on each Monday, another one on each Tuesday, not appear in any other A manuscript transmitting the same days.9 Thus, the large and so on. These vespertini follow the Psalter ordering, although, again, not strictly, number of unica vespertini in tradition A’s Lent may reflect the particular richness suggesting links between Tuesdays and Thursdays (instead of between Wednesdays of E-L Ms 8’s repertoire. and Thursdays, as in tradition A). Vespertini in B’s Lent have a laudatory character and only deal with the topic of the God/light and sin/darkness duality. Thus, while The A and B Lent and Eastertide temporale through their vespertini10 tradition A vespertini and other tradition B genres do follow the differentiation of The vespertinus repertoire that tradition A’s E-L Ms 8 gives for the temporale offices the two halves of Lent, the tradition B vespertini do not. Finally, instead of a long of Lent and of the beginning of Eastertide is presented in table 1 (p. 304). One can see Laudaberunt te containing a list, tradition B (in E-Tc 35.5) has a modest version of this vespertinus for Easter Sunday (with a main section and one verse only), and an incipit for this chant on the Easter octave. 9. Vespertini A solis ortu and Tenebrae non obscurantur (for the quotidian days, i.e. for Christmastide and Post-Pentecost time), Sicut Christus surgens (for the Holy Cross, falling in Eastertide), and Vespertina oratio ascendat (for the Friday of Apostolic Litanies) are only found in E-L Ms 8, not existing in any of the other A manuscripts. 11. Previous authors have noted the division of Old Hispanic Lent into two halves. See, e.g., Pinell (1998, p. 289). 10. As mentioned in note 5 above, the sanctorale vespertini in E-Tm n.i. 1325 and their cognates in tradition A manuscripts were not included in my sample of contrasted chants because the former lack musical notation. 12. More details in Rojo Carrillo (2017).

296 297 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

Table 3 (p. 308) displays the A and B vespertinus assignments. The five vespertini notes of the version in the second manuscript are contained in the version of the preceded by a cross (+), appear in both A and B’s Lent. Four of them were sung first manuscript (cmr2). Finally, I obtained the average of the shared readings by repeatedly throughout Lent. The only vespertini that have the same assignments in following the operation: (cmr1+cmr2)/2. both traditions are Inlumina domine vultum for Palm Sunday, and Laudaberunt te domine for Easter Sunday and Easter octave weekdays. Although the chant melodies When comparing the melody given to Usque ad vesperum by tradition A’s E-L Ms given by each tradition are not identical, their melodic contours and patterns are 8 and tradition B’s E-Tc 35.5 (example 1, p. 300), it can be seen that they share the related. Chants in italics are unica. In tradition B there is only one vespertinus, Fecit same final ending pattern (over domino‘ ’). lunam, that does not appear elsewhere. However, it has the same text that is used in a verse of Laudaberunt te domine, which in both traditions was sung several times Most of the rest of the chant is exactly the same in both manuscripts, reaching during Eastertide. Thus, this B vespertinus is not as unique as it seems. Chants eighty-eight percent of shared melodic reading. Another tradition A manuscript, preceded by an asterisk also appear with non-Lenten assignments in E-L Ms 8 and/ the eleventh-century liber ordinum et misticus E-SI Ms 3, has this same vespertinus or in other A manuscripts. All B vespertini (except Fecit) appear in A manuscripts, assigned to a Quotidian Monday. Therefore, A manuscripts also assign part of 15 where they present similar melodic contours, and are assigned to quotidian offices their shared material differently. If we compare this other A melody of Usque ad and to votive offices. vesperum with the one in E-L Ms 8, it can be seen that they also coincide in most of their melodic readings, reaching eighty-six percent similarity. However, this The shared material similarity is lower than that between the E-L Ms 8 and the B manuscript. If we As part of my Ph. D. research (2017)13 I analysed the texts and melodies of all compare the E-SI Ms 3 version with the one in tradition B, we also find coincidences, extant vespertini. Among other outcomes, the preliminary results of my analysis but fewer, with only seventy-eight percent of melodic readings in common. Note allowed me to assess whether the melodies also reflect two different traditions. I that despite the shared material, this other A manuscript adds one more note to the have chosen two of the shared chants to give an idea of the A-B relationship in this final pattern (on ‘domino’), which makes it different from the other two versions in specific genre. In these chants we can see that the texts in A and B were selected the most important part of the chant. The number of notes in each version also show and adapted in exactly the same way. The melodies, instead, are slightly different. that the Lent versions in E-L Ms 8 and E-Tc 35.5 are close, presenting fewer notes I will, therefore, focus on the results of the melodic analysis. In my analysis of the than the E-SI Ms 3 Quotidian version. vespertini melodies, I examined their melodic readings, recurrent patterns, and number of notes per chant.14 Sometimes the totality of a particular melodic reading Thus, tradition A’s E-L Ms 8 version is closer to the tradition B version (in E-Tc 35.5) can be found in another, more ornate, cognate of the same chant. In this case, I than to the other A version (in E-SI Ms 3), in both melodic reading and number of understand each instance as a version of the same melody. However, when trying to notes. The version in tradition A’s E-SI Ms 3 is closer to the version of E-L Ms 8 (also detect the common points between the ornamented version of a particular melody tradition A) in melodic reading, but it is closer to the tradition B version in number of and a simpler version, we could in some cases find that less than fifty percent of the notes. These results place the versions for the same time of the year, Lent, in a closer material of the ornamented version appears in the sober one. However, this does not musical relationship, despite their different liturgical tradition tags. They also show mean that these versions are two different melodies. To avoid such inaccuracies, that the melodic differences between the vespertini shared by traditions A and B are when contrasting two versions of the same chant (with a shared melodic reading not always more marked than the melodic differences that can be found between the over the same text), I compared the number of common notes with the total number vespertini shared by manuscripts of the same tradition (tradition A in the previous of notes in each version. I did this by considering two numbers: one expressing example). how many notes of the version in the first manuscript are contained in the version given in the second manuscript (cmr1); and the other number expressing how many I use the main section of a vespertinus with the verses, Sacrificate sacrificium, as my second example (example 2, p. 300).

13. This dissertation contains a detailed list of the liturgical assignments of the vespertini. For lists of the liturgical assignments of other Old Hispanic chant genres see Randel (1973).

14. With the term ‘melodic reading’ I am referring to the relative pitch-direction of each of the notes that are 15. Another notated Usque ad vesperum is preserved in a manuscript considered to belong to tradition A, the liber contained in each neume. This is expressed through sigla (see example 1). This transcription system is based on misticus GB-Lbl Add MS 30845. Here it receives yet another liturgical assignment: the third day of Canonical Litanies. that proposed by Hornby and Maloy (2012), and employed by these same scholars in their most recent works, and by This manuscript, as E-SI Ms 3, has been ascribed to the Abbey of Santo Domingo de Silos, in northern Spain, but the Chant Editing and Analysis Program (neumes.org.uk/editor) which I have developed together with a team led by scholars disagree about its copy-date (see http://musicahispanica.eu/source/20196). I have chosen E-SI Ms 3 for the Hornby at University of Bristol. examples in this paper, because it presents a copy-date, 1039, in its fol. 177.

298 299 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

Example 1. Comparison of the melodies in different renditions of vespertinus Usque ad vesperum

Manuscript Usque ad vesperum quam magnificata

E-L Ms8 N-N N N-NS-NUUHLHL N NS-NHL-NH-NHL+NL-NHLH E-L Ms 8, fol. 118 E-Tc 35.5 N-N N N-NHL-NH+NHHL N NHL-NLH-NH-NHL+NL-NHLH

E-SI Ms 3 N-N N N-NS-NUS+NHL+NS+NHL N NS-NHL-NH-NHL+NL-NHLH

sunt opera tua domine Total notes

E-L Ms8 N NHLH-NL-N NH-N NHL+NL-NHLH-N 51

E-Tc 35.5 N NHLH-NL-N NLH-NH NHL+NL-NHLH-N 54

E-SI Ms 3 N NHLH-NL+NL-N NHHL-NH NHL+NL-NLHLH-N 61

Example 2. Comparison of melodies for the first section (respond) of vespertinus Sacrificate sacrificium E-Tc 35.5, fol. 74v

Manuscript Sacrificate sacrificium

E-L Ms8 NSH-N-N-NH+NL-NHHLH+NH+NHL NH-NH-N-NLHH-N

E-Tc 35.5 N-N-N-N+NUL-NHL+NHHL NH-NH-NHH-N-N

E-SI Ms 3 NS-N-N-NH+NLH-NH+NLHL NH-NH-N-NLHH-N E-Tc 35.5, fol. 74v vespertinum et sperate

E-L Ms8 NUUH-NHL-NH+NLH+NHH+NHLH-NH NH NHH-NHL-NL

E-Tc 35.5 NHHH-NL-NLH+NHH-N N NHHL-NHLH-NL

E-SI Ms 3 NUUUUHLH-NHL-NH+NLH+NHH+NHL-N N NHH-NHL-NL

filii hominum in domino Total notes

E-L Ms8 NUUHL-NS NHL-NHHLH-NH NHH NHHLLLHLH-N 90

E-Tc 35.5 NHHL-NHH NH-NLH-NH NHL N-NH+NHLH-N 72

E-SI Ms 3 NUUUHL-NS NHL-NHHLH-NH NHL N-NHHLL+NHL-N 88

E-L Ms 8, fol. 281v Legenda N Neutral – unknown – relative pitch height L a lower pitch than the note immediately preceding it in the same neume H a higher pitch than the note immediately preceding it in the same neume S the same pitch as that of the note immediately preceding it in the same neume D either the same or a lower pitch than the note immediately preceding it in the same neume U either the same or a higher pitch than the note immediately preceding it in the same neume Any of italicised indicate that they have a neume or pen-stroke regarded as quilismatic above in previous scholarship E-SI Ms 3, fol. 148 - To differentiate syllables of each word and the melodic readings of each of these syllables

+ To separate the different neumes within a syllable E-Tc 35.5, fol. 37

300 301 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

In tradition B, this chant is assigned to some Sundays of Lent; in tradition A it is for Bibliography Quotidian Sundays.16 The text is again exactly the same in all manuscripts, despite being a modified version of part of Psalm 4:6. E-L Ms 8 and E-Tc 35.5 have versions that Antiphonarium mozarabicum de la catedral de León (1928), Burgos: Imprenta Aldecoa. share seventy-seven percent of their melodic readings and present the same pattern at Brou Louis (1954), ‘Le joyau des antiphonaires latins. Le manuscrit 8 des Archives de la Cathédrale de León’, the end of the chant (‘domino’). In this case, however, E-L Ms 8 and E-SI Ms 3 share Archivos Leoneses, Revista del Centro de Estudios e Investigación de ‘San Isidoro’, 8/15-16, pp. 7-114. more notes, reaching eighty-eight percent of common melodic readings. Now the E-SI Díaz y Díaz Manuel C. (2007), Some Incidental Notes on Music Manuscripts in Susana Zapke ed., Hispania Ms 3 version and the tradition B version present seventy percent of shared melodic Vetus: Musical-Liturgical Manuscripts from Visigothic Origins to the Franco/Roman Transition (9th- 12th Centuries), Bilbao: Fundación Bbva, pp. 93-111. readings, the lowest percentage of similarity. The notes per chant also show a stronger link between the A manuscripts, with B closer to E-SI Ms 3 than to E-L Ms 8. It is worth Gutiérrez Carmen Julia et al. (2016-), Spanish Early Music Manuscripts Database, Madrid: Universidad Complutense de Madrid (http://musicahispanica.eu). emphasising that despite the closer relationship between the two A manuscripts, E-SI Ms 3 uses a different melodic gesture at its final cadence, a fundamental section of any Hornby Emma et al. (2013-), Chant Editing and Analysis Program, Bristol: University of Bristol (neumes. org.uk/editor). chant. The results of this example present an A-B relationship that is different from that of the previous example, but both examples coincide in giving different melodic Hornby Emma - Maloy Rebecca (2012), Toward a Methodology for Analyzing the Old Hispanic Responsories in Robert Klugseder ed., Cantus Planus: Study Group of the International Musicological Society. densities to different seasons: fewer notes for Lent, and more for Quotidian chants. Papers read at the 16th meeting, Vienna, Austria, August 2011, Pukersdorf: Hollinek, 2012, pp. 242-249. They also coincide in showing that the melodic relationship between versions of a – (2013), Music and Meaning in Old Hispanic Lenten Chants: Psalmi, threni and the Easter Vigil given chant reflects their liturgical assignments:the more similar these assignments, Canticles, Woodbridge, Suffolk: Boydell & Brewer. the closer their melodies are. Janini José (1977), Introducción in José Janini - Ramón Gonzálvez - Anscanio M. Mundó, Manuscritos litúrgicos de la Catedral de Toledo, Toledo: Diputación Provincial (Publicaciones del Instituto Preliminary conclusions Provincial de Investigaciones y Estudios Toledanos, Serie Estudios, Catálogos, Repertorios), pp. 9-49.

The examples provided here are only two of the many that can be found in the Millares Carlo Augustin – Manuel Cecilio Díaz y Díaz - Anscari Manuel Mundó Marcet - José Manuel vespertinus repertoire. The musical similarities between the chants of this genre Ruiz Asencio - Blas Casado Quintanilla - Enrique Lecuona Ribot eds. (1999), Corpus de códices visigóticos, Las Palmas de Gran Canaria: Universidad de Educación a Distancia, Centro Asociado de seem to be determined more by their liturgical assignments and by their particular Las Palmas de Gran Canaria. manuscript than by their scholarly-coined tradition A/B tag. The liturgical differences Mundó Anscari M. (1965), ‘La datación de los códices litúrgicos visigóticos toledanos’, Hispania Sacra between the tenth-century A antiphonary (E-L Ms 8) and the late thirteenth- and early 18/35, pp. 1-25. fourteenth-century B manuscripts (E-Tc 35.5 and E-Mn MSS/10110) might be owed to inell Los textos de la antigua liturgia hispánica. Fuentes para su estudio ivera the fact that they were produced in very different contexts. This hypothesis becomes P Jordi (1965), in Juan F. R Recio ed., Estudios sobre la liturgia mozárabe, Toledo: Diputación Provincial (Publicaciones del stronger if we take into consideration that E-L Ms 8 and, not only E-SI Ms 3, but also Instituto Provincial de Investigaciones y Estudios Toledanos, s. 3), pp. 109-164. all of the other tradition A manuscripts were copied in contexts that were closer to – (1972), Liturgia hispánica in Jose Vives Gatel - Quintin Aldea Vaquero - Tomas Marin Martinez each other than to B, and, nonetheless, also present unique chants and a number of eds., Diccionario de la Historia Eclesiástica de España, 4 vols., Madrid: Instituto Enrique Flórez. different assignments and melodic particularities in their shared repertoire. Moreover, – (1978), El problema de las dos tradiciones del antiguo rito hispánico […] in Liturgia y música mozárabes: the substantial number of vespertini that only appear in tradition A’s Lent might be a Ponencias y comunicaciones presentadas al I Congreso Internacional de Estudios Mozárabes, Toledo reflection of the particular richness of the manuscript that preserves them, E-L Ms 8. 1975, Toledo: Instituto de Estudios Visigótico-Mozárabes de San Eugenio, pp. 3-44. Additionally, the fact that B coincides with the antiphonary and other A manuscripts – (1998), Liturgia Hispánica, Barcelona: Biblioteca Litúrgica. in its very few Eastertide vespertinus assignments calls into question the application Randel Don M. (1969a), The Responsorial Tones for the Mozarabic Office, Princeton, NJ: Princeton University of the A/B division to the entire liturgical year, and invites speculation as to whether Press, pp. 93-100. 17 such a division should be restricted to a particular season, Lent. – (1969b), ‘Responsorial Psalmody in the Mozarabic Rite’, Études gregoriéns 10, pp. 87-116.

– (1973), An Index to the Chant of the Mozarabic Rite, Princeton NJ: Princeton University Press.

16. Notated versions of Sacrificate sacrificiumalso exist in tradition A manuscripts E-SI Ms 6, GB-Lbl Add MS 30851, E-Tc Rojo Carrillo Raquel (2017), Text, liturgy and music in the Old Hispanic rite: the vespertinus genre, Ph. D. 35.4 and F-Pn NAL 2199, assigned to Quotidian Sundays; GB-Lbl Add MS 30845, assigned to the office of Primitis (Harvest dissertation, University of Bristol. office); and E-Zfm Ms 418, for the Sunday before Carnes Tollendas. E-L Ms 8 also assigns it through incipits to the Primitis and the Wedding offices. – (2018, forthcoming), Old Hispanic Chant Manuscripts of Toledo: Testimonies of a Local or of a Wider 17. I gratefully acknowledge the European Research Council, which, through its Old Hispanic Office Project led by Emma Tradition? in Yasmin Beale-Rivaya - Jason Busic, eds., A Companion to Medieval Toledo: Shared Hornby at University of Bristol, granted me academic and financial support allowing me to write this article. Common Spaces: Medieval Toledo 711-1517, Boston: Brill.

302 303 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

Appendix Third Sunday of Lent – Principes persecuti sunt me gratis Ps 118:161 Mediante, second vespers et a verbis tuis formidabit cor meum Table 1. Antiphonary of León (E-L Ms 8), temporale vespertini of Lent and Eastertide (original spelling; repetition cues in italics; VR = verse) Monday, only vespers Domine libera animam meam a labiis Ps 119:2 at the end of the day iniquis et a lingua dolosa Lent Tuesday, only vespers Levavi oculos meos ad montes unde veniet Ps 120:1 Assignment Vespertini Biblical source at the end of the day auxilium mihi Carnes Tollendas, Elevatio manuum mearum sacrificium Ps 140:2 Wednesday, only vespers Eripe me domine ab homine malo et a viro Ps 139:2 first vespers vespertinum at the end of the day iniquo libera me VR Dirigatur domine oratio mea in conspectu Ps 140:2 Thursday, only vespers Conserva me domine de manu peccatoris Ps 139:5 tuo sicut incensum ves […] at the end of the day ab hominibus iniquis livera me Carnes Tollendas, Psallam et intellegam in via inmaculata Ps 100:1-2 Friday, only vespers Expandi manus meas ad te velociter Ps 142:6-7 second vespers quando venies ad me at the end of the day exaudi me domine VR Misericordiam et iudicium cantabo tibi Ps 100:1 Fourth Sunday of Lent, Inlumina oculos meos domine ne umquam Ps 12:4 domine quando […] first vespers obdormiam in morte Monday, only vespers Videbo caelos domine opera digitorum Ps 8:4 VR Nequando dicat subplantator meus Ps 12:5 at the end of the day tuorum lunam et stellas quas tu fundasti prevalui adversus eum in morte Tuesday, only vespers Quoniam tu inluminas lucernam meam Ps 17:29 Fifth Sunday of Lent - Inlumina domine vultum tuum super Ps 118:135 at the end of the day domine deus meus inlumina tenebras meas Palm Sunday, first vespers serbos tuos et doce nos iustificationes tuas Wednesday, only vespers Inluminatio mea deus defensor vite meae Ps 26:1 VR Declaratio sermonum tuorum inluminet Ps 118:130 at the end of the day a quo trepidabo nos et intellectum dat parvulis et do […] Thursday, only vespers Inluminatio mea et salus mea dominus Ps 26:1 Holy Wednesday, Inlumina oculos meos domine ne umquam Ps 12:4 at the end of the day only vespers at the end obdormiam in morte of the day Friday, only vespers Vespere mane et meridie te laudamus Ps 54:18 VR Respice et exaudi me domine deus meus Ps 12:4-5 at the end of the day domine nequando dicat inimicus meus prevalui adversus eum in morte […] First Sunday of Lent, Aput te est domine fons vite et in lumine Ps 35:10 first vespers tuo videbimus lumen VR Filii hominum in protectione alarum Ps 35:8-9 Eastertide tuarum sperabunt inebriabuntur ab uberibus Easter Sunday, Laudaberunt te domine omnes angeli tui Ps 73:16 domus tue et torrentem volumtatis tue potum second vespers, et hymnum dixerunt dum perficeres dabis eis domine fons vite […] and Eastertide weekdays, fabricam celi tu est dies et tua est nox First Sunday of Lent, Usque ad vesperum quam magnificata sunt Ps 103:23-24 first vespers tu fecisti solem et lunam second vespers opera tua domine VR Tu confirmasti in virtute tua mare tu Ps 73:13 Monday, only vespers Dominus inluminatio mea Ps 26:1 contribulasti capita draconum super aquas at the end of the day tu fe […]

Tuesday, only vespers Vespertina oratio ascendat ad te domine Ps 140:2 II Fecit lunam in tempore suo sol agnobit Ps 103:19-20 at the end of the day occasum suum posuisti domine tenebras et Second Sunday of Lent, Exaudi nos deus salvator noster Ps 64:6 facta est nox first vespers III Tu confregisti caput draconis magni Ps 73:14 VR Preparans montes in fortitudine tua Ps 64:7 exaltatus in magna potentia salva […] et dedisti eum in esca populo eziopo tu Third Sunday of Lent – Dominus inluminatio mea et salus mea Ps 26:1 IIII Tu fecisti omnes terminos terre estatem Ps 73:17 Mediante, first vespers quem timebo et ver tu fecis VR Ipse est defensor vite meae dum Ps 26:1-2 V Tu disrupisti fontes et torrentes tu exicasti Ps 73:15 adpropinquant nocentes quem […] flumina etan tu f […]

304 305 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

Table 2. Libri mistici of Toledo (E-Tc 35.5 and E-Mn MSS/10110) temporale vespertini of Lent and Eastertide (original spelling; repetition cues in italics; VR = verse)

Lent (E-Tc 35.5 and E-Mn MSS/10110) Assignment Vespertini Biblical source Monday to Friday same as first week

First Sunday (called Carnes In noctibus extollite manus vestras in Ps 133:2 Fifth Sunday (called fourth Sacrificate sacrificium vespertinum et Ps 4:6 Tollendas in tradition A), sancta et benedicite domino Sunday in tradition A) sperate filii hominum in domino second vespers (probably first vespers employed for all second vespers VR Confitemini Domino et invocate of the first half of Lent) nomen eius et spe […] Monday, only vespers Tu inluminas lucernam meam domine Ps 17:29 Fifth Sunday (called fourth Usque ad vesperum quam magnificata Ps 103:23-24 at the end of the day Sunday in tradition A), sunt opera tua domine second vespers Tuesday, only vespers Inluminatio mea et salus mea dominus Ps 26:1 at the end of the day Monday to Friday same as first week

Wednesday, only vespers Vespere mane et meridie te laudamus Ps 54:18 Palm Sunday, Inlumina domine vultum tuum super Ps 118:135 at the end of the day domine first vespers servos tuos et doce eos iustificationes tuas

Thursday, only vespers Dominus inluminatio mea et salus mea Ps 26:1 Declaratio sermonum tuorum inluminet Ps 118:130 at the end of the day quem timebo nos et intellectum dat parvulis et do […]

Friday, only vespers Sol agnouit occasum suum, posuisti Ps 103:19 Palm Sunday, Usque ad vesperum quam magnificata Ps 103:23-24 at the end of the day tenebras et facta est nox second vespers sunt opera tua dmine

Second Sunday (called Fecit lunam in tempora sol agnovit Ps 103:19 Holy Monday to Wednesday same as first week first Sunday in tradition A) occasum suum first vespers VR Exiet homo ad opus suum et ad Ps 103:23 Easter Sunday and Eastertide (E-Tc 35.5) operationem suam usque ad vesperum Easter Sunday, Laudaberunt te domine omnes angeli Ps 73:16 Monday to Friday same as first week second vespers, tui et hymnum dixerunt dum perficeres and Eastertide weekday, fabricam celi tu est dies et tua est nox Third Sunday (called Sacrificate sacrificium vespertinum et Ps 4:6 first vespers tu fecisti solem et lunam second Sunday in tradition A) sperate filii hominum in domino first vespers Tu confirmasti in virtute tua mare tu Ps 73:13 VR Confitemini Domino et invocate nomen Ps 104:1 contribulasti capita draconum super aquas eius et spe […] Tu fecisti […] Third Sunday (called In noctibus extollite manus vestras in Ps 133:2 second Sunday in tradition A) sancta et benedicite domino second vespers Monday to Friday same as first week

Fourth Sunday (called Fecit lunam in tempora sol agnovit Ps 103:19, 23 third Sunday in tradition A) occasum suum first vespers VR Exiet homo ad opus suum et ad Ps 103:23 operationem suam usque ad vesperum Fourth Sunday (called Usque ad vesperum quam magnificata Ps 103:23-24 third Sunday in tradition A) sunt opera tua domine second vespers

306 307 CANTUS PLANUS 2014 ROJO CARRILLO — THE VESPERTINI AS WITNESSES TO THE TRANSMISSION OF OLD HISPANIC CHANT

Table 3. Comparison between tradition A (E-L Ms 8) and tradition B (E-Tc 35.5 and E-Mn MSS/10110) Vespertini assignments in temporale offices of Lent and Eastertide, original spelling Vespertini in bold: same assignment in both traditions; vespertini in italics: unica Vespertini preceded by an *asterisk also appear in E-L Ms 8 or other tradition A manuscripts with different liturgical assignments Vespertini preceded by a + appear in Lent in both traditions

Lent vespers service Vespertini tradition A Vespertini tradition B Mediante or Third (A)/Fourth (B) +Dominus inluminatio mea Fecit lunam in tempora (biblical source) (biblical source) Sunday of Lent, first vespers (Ps 26:1-2, different melody (Ps 103:19 and 24) and length) In Carnes Tollendas (A)/First *Elevatio manuum mearum none preserved Sunday of Lent (B), first vespers (Ps 140:2) Mediante or Third (A)/Fourth (B) Principes persecuti sunt +Usque ad vesperum Sunday of Lent, second vespers (Ps 118:161) (Ps 103:23-24) In Carnes Tollendas(A)/First Psallam et intellegam *In noctibus extollite Sunday of Lent (B), second vespers (Ps 100:1-2) (Ps 133:2) Monday, fourth week of Lent, Domine libera animam same as Monday, only vespers at the end of the day (Ps 119:2) first week Monday, first week of Lent, Videbo caelos domine *Tu inluminas lucernam only vespers at the end of the day (Ps 8:4) (Ps 17:29) Tuesday, fourth week of Lent, Levavi oculos meos same as Tuesday, only vespers at the end of the day (Ps 120:1) first week Tuesday, first week of Lent, Quoniam tu inluminas +Inluminatio mea et salus only vespers at the end of the day lucernam (Ps 17:29) (Ps 26:1) Wednesday, fourth week of Lent, Eripe me domine same as Wednesday, only vespers end of the day (Ps 139:2) first week Wednesday, first week of Lent, Inluminatio mea deus +Vespere mane et meridie only vespers at the end of the day (Ps 26:1) (Ps 54:18) Thursday, fourth week of Lent, Conserva me domine same as Thursday, only vespers at the end of the day (Ps 139:5) first week Thursday, first week of Lent, +Inluminatio mea et salus +Dominus inluminatio mea only vespers at the end of the day (Ps 26:1) (Ps 26:1) Friday, fourth week of Lent, Expandi manus meas same as Friday, only vespers at the end of the day (Ps 142:6-7) first week Friday, first week of Lent, +Vespere mane et meridie *Sol agnouit occasum only vespers at the end of the day (Ps 54:18) (Ps 103:19) Fourth (A)/Fifth (B) Sunday Inlumina oculos meos *Sacrificate sacrificium of Lent, first vespers (Ps 12:4-5) vespertinus (Ps 4:6) First (A)/Second (B) Sunday Aput te est Fecit lunam in tempora of Lent, first vespers (Ps 35:8-10) (Ps 103:19 and 23) Fourth (A)/Fifth (B) Sunday no vespertinus +Usque ad vesperum of Lent, second vespers (Ps 103:23-24) First (A)/Second (B) Sunday +Usque ad vesperum no vespertinus of Lent, second vespers (Ps 103:23-24) Monday to Friday, fifth week same as Monday to Friday, same as Monday to Friday, of Lent fourth week first week Monday, second week of Lent, +Dominus inluminatio mea same as Monday, only vespers at the end of the day (Ps 26:1) first week Palm Sunday, first vespers Inlumina domine vultum Inlumina domine vultum (Ps 118:130, 135) (Ps 118:130, 135) Tuesday, second week of Lent, *Vespertina oratio ascendat same as Tuesday, only vespers at the end of the day (Ps 140:2) first week Palm Sunday, second vespers no vespertinus +Usque ad vesperum (Ps 103:23-24) Wednesday to Friday, second same as Wednesday, same as Wednesday, week of Lent, only vespers at the Thursday and Friday, Thursday and Friday, Holy Monday and Tuesday, same as Monday and same as Monday and end of the day first week first week sixth week of Lent Tuesday, fourth week Tuesday, first week

Second (A)/Third (B) Sunday *Exaudi nos deus *Sacrificate sacrificium Holy Wednesday, Inlumina oculos meos same as Wednesday, of Lent, first vespers (Ps 64:6-7) (Ps 4:6 and Ps 104:1) sixth week of Lent, (Ps 12:4-5, different melody first week only vespers end of the day and length) Second (A)/Third (B) Sunday no vespertinus *In noctibus extollite of Lent, second vespers (Ps 133:2) Eastertide Monday to Friday, same as Monday & Tuesday same as Monday to Friday, third week of Lent second week and Wednesday, first week Easter Sunday, second vespers, Laudaberunt te domine Laudaberunt te domine Thursday and Friday, first and Eastertide weekday, first (Ps 73:16 in main section (Ps 73:16 in main section week vespers and first verse; Ps 103:19-20, and first verse) 73:14, 73:17, 73:15 in verses II to IIII)

308 309 Hanna Zühlke

Hirsau in Norditalien: Zur liturgischen Musikpraxis des Benediktinerklosters Moggio im zwölften und dreizehnten Jahrhundert

«Hic incipitur ordo diuini officii per circulum anni secundum Hirsiacenses»1 – Die Rubrik, die dem Normtext des friulanischen Benediktinerklosters Moggio voran- steht, macht unmissverständlich deutlich, auf welche liturgische Tradition die nach- folgenden Bestimmungen für Messe und Offizium rekurrieren. Als eine von drei norditalienischen Abteien gehörte Moggio im zwölften Jahrhundert zum Hirsauer Reform­verband, der neben Klöstern im deutschen Südwesten, also in unmittelbarer Nähe zum Reformzentrum Hirsau, auch weiter entfernte Filiationen einschloss: Über Subzentren der Reform wurde die Hirsauer Bewegung bis nach Stolpe im heutigen Mecklenburg getragen; sie erreichte die Donauklöster Göttweig und Melk im Osten und drang in den südlichen Alpenraum bis ins Friaul vor.2

Wann und über welches Kloster Moggio dem Hirsauer Kreis beitrat, ist – anders als im Falle der benachbarten Benediktinerabteien Rosazzo und Santa Maria di Aquileia, die wohl über Millstatt hirsauisch geprägt wurden (Jakobs 1961, pp. 45-46; Scalon 1984, pp. 57-59; Baum 2002, p. 154) – nicht zu ermitteln. Die aus Moggio überlieferten ältesten Urkunden, die Reinhard Härtel (1985) einer eingehenden Analyse unterzogen hat, las- sen die Herkunft der ersten Mönche, die das Kloster mit dessen Weihe im Jahr 1118 oder 1119 besiedelten, offen.3 Dass sich die Gründungsinitiative des Patriarchen Ulrich i. von Aquileia (1086-1121), der gleichzeitig als Abt im schweizerischen Sankt Gallen wirkte,4 auch auf die Berufung von Mönchen aus seiner Heimatabtei erstreckte, wird in der Klosterlegende Moggios zwar erzählt,5 kann anhand der Quellen jedoch nicht belegt werden (Ochsenbein 1993, pp. 316-318). Denkbar ist also gleichermaßen eine Übernahme der Hirsauer Observanz bereits zum Zeitpunkt der Gründung Moggios.

Die liturgischen Handschriften aus dem Bestand der ehemaligen Klosterbibliothek, heute zum größten Teil in der Biblioteca Arcivescovile di Udine und in der Bodleian

1. GB-Ob MS. Canon. Liturg. 325, fol. 15v.

2. Zur Ausbreitung der Hirsauer Reform vgl. grundlegend Jakobs (1961).

3. Zur nicht mehr vollständig rekonstruierbaren Frühgeschichte Moggios vgl. auch Härtel (1994; 2001, pp. 832-836), ferner Menis (1994a).

4. Zu Ulrich vgl. ausführlich Vogler (1994).

5. Diese ist überliefert in Form einer Inschrift aus dem späten achtzehnten oder neunzehnten Jahrhundert, die sich im Inneren der Klosterkirche in Moggio befindet. Sie ist zitiert bei Ochsenbein (1993, p. 315): «Ex celeberrimo monasterio S. Galli Helvetiorum rogatu Woldarici patrirachae [sic] illius olim coenobii clavum tenentis xii monachi profecti novum Mosacense monasterium cum Bebolpho primo abbate an. mcxix solemniter ingressi sunt […]».

311 CANTUS PLANUS 2014 ZÜHLKE — HIRSAU IN NORDITALIEN

Library in Oxford aufbewahrt,6 helfen in dieser Hinsicht nicht weiter: Für die wenigen gestellt: «Die Abtei scheint über kein wirkliches scriptorium verfügt zu haben; für die Quellen, die vor 1118/1119 datieren, sind verschiedene Herkunftsräume auszumachen. Versorgung mit den für den Gottesdienst nötigen Büchern wandte man sich an die In der Forschung wurden etwa Lokalisierungen in den Salzburger und Regensburger Abteien jenseits der Alpen […]» (Scalon 1987, p. 624). Die Zuweisung einzelner Quel- Raum vorgenommen.7 Handschriften Sankt Galler Provenienz finden sich darunter len an einen konkreten Herstellungs- oder Bestimmungsort erfordert im Kontext des nicht, wie Peter Ochsenbein (1993, p. 320) herausgearbeitet hat. Zudem ist unklar, Hirsauer Reformkreises ein methodisch reflek­tiertes Vorgehen: Nicht zwingend muss wann und auf welchem Weg diese Codices nach Moggio gelangt sind. Die Überlie- hier die inhaltliche Übereinstimmung zwischen Handschriften deren gleiche Herkunft ferung hirsauisch geprägter Handschriften aus dem friulanischen Kloster setzt zwar andeuten, sondern die Gemeinsamkeiten können zunächst allein durch die Hirsauer erst um die Mitte des zwölften Jahrhunderts ein,8 gleichwohl lässt sich daraus nicht Prägung der verschiedenen Herkunftsorte bedingt sein. So konnte Felix Heinzer (1992, automatisch ein verspäteter Anschluss Moggios an die Reformbewegung ableiten. pp. 322-334) auf der Basis seiner Forschungen zur Hirsauer Liturgie mehrere der von Scalon vorgenommenen Herkunftsbestimmungen zu den Moggio-Handschriften im Den Hirsauer Spuren in dem aus Moggio erhaltenen hochmittelalterlichen Quellen­ Udineser Bestand korrigieren. Aufgrund von Konkordanzen im Bereich der Responso- bestand ist erstmals Cesare Scalon in seiner 1979 publizierten grundlegenden Arbeit rien des Totenoffiziums und auffälliger Parallelen in der Allerheiligenlitanei zwischen zu den Handschriften der Udineser Bibliothek nachgegangen. Dabei wurden zwei den Codices der Biblioteca Arcivescovile (I-UDba) 72, 73 und 77 und Handschriften aus Dinge deutlich: Zum einen weisen Textschrift, Buchschmuck und musikalische No- Sankt Emmeram in Regensburg war Scalon (1979, pp. 28-32, 136-140, 143-147) davon tation der Codices des zwölften und dreizehnten Jahrhunderts grundsätzlich in den ausgegangen, dass erstere dort entstanden sein müssten. Dies wiederum brachte ihn 9 süddeutsch-österreichischen Raum. Zum anderen legen die paläographischen und zu der Annahme, dass Moggio über Sankt Emmeram Anschluss an Hirsau gefunden inhaltlichen Befunde die Herkunft der Handschriften aus verschiedenen Hirsauer hat.11 Heinzers Untersuchungen (1992, pp. 323-330) haben jedoch gezeigt, dass man Klöstern nahe. Ähnliche Ergebnisse erzielte Rudolf Flotzinger (1991) im Zuge seiner im Reformkreis eine einheitliche Responsorienreihe für das Totenoffizium und eine Untersuchung der Choralhandschriften österreichischer Provenienz in der Bodleian gemeinsame Allerheiligenlitanei verwendete. Scalons Provenienzbestimmung war 10 Library/Oxford. Angesichts der heterogenen Zusammensetzung des friulanischen damit die Grundlage entzogen. Quellenkorpus hat Scalon die Existenz eines eigenen Skriptoriums in Moggio in Frage Die seit Heinzers Wiederentdeckung des Hirsauer Liber ordinarius erkennbar gewor- 6. Einen Überblick geben Foligno (1913), Casarsa (1968), Scalon (1979, pp. 20-33) und Flotzinger (1991, pp. 38-66). dene Hirsauer Liturgie und im Speziellen die von Hirsau aus verbreiteten liturgischen 12 7. Salzburg: I-UDba Ms. 4, ein Homiliar aus dem ersten Viertel des neunten Jahrhunderts; vgl. dazu Scalon (1979, Gesänge sind in Handschriften aus verschiedenen Tochterklöstern der Reform ganz pp. 70-76) und Bischoff (1980, p. 148). Regensburg: I-UDba Ms. 234, ein Graduale mit Tropen und Kyriale aus dem späten elften Jahrhundert. Scalon (1979, pp. 30, 230-231) bringt die Handschrift aufgrund ihrer nachpfingstlichen Alleluia-Verse unterschiedlich kontextualisiert: Der Beitritt zum Hirsauer Verband bedeutete nicht mit Sankt Emmeram in Verbindung. Felix Heinzer (1992, pp. 331, 333) weist dagegen auf die Unterschiede zur Tradition unweigerlich auch den Verzicht auf die Ausbildung eines eigenen musikalisch-litur- des Regensburger Benediktinerklosters hin und identifiziert in der Alleluia-Reihe der Udineser Handschrift Regensburger Diözesanusus. gischen Profils, die Preisgabe bereits bestehender lokalliturgischer Traditionen oder 8. Ob es sich bei I-UDba Ms. 78, einem um 1100 angefertigten Graduale-Sakramentar bislang ungeklärter Provenienz, aber die Ablehnung neuer Entwicklungen im Bereich der Mess- und Offiziumslitur- um eine Handschrift aus dem Reformkreis handelt, ist fraglich. Sequentiar und Tropar entbehren der für Hirsau typischen Repertoirekonstellation; zu dieser vgl. Haug (1994) und Kruckenberg (1999). Die in der Literatur vorgenommenen gie. Handschriften aus dem Reformkreis lassen oft ganz eigene Prinzipien bei der Aus- Lokalisierungen des Codex nach Weingarten (Huglo 1976, p. 318; Scalon 1979, pp. 32, 147-148) und Melk (Heinzer wahl von Gesängen erkennen, die nicht selten lokale, regionale und überregionale 1997, p. 160*; Camilot-Oswald 1997, p. lxxxiv) halten einer genaueren Überprüfung nicht stand: Zwar lässt sich das im Kalendarium (fol. 4) nachgetragene Kirchweihdatum auf Weingarten beziehen (die zweite Abteikirche wurde dort am 12. Traditionen zusammenführen. Erst die Aufdeckung des je spezifischen Gefüges aus November 1182 geweiht), jedoch sprechen keine weiteren inhaltlichen Indizien für das oberschwäbische Kloster. Eine Konkordanz zur nachpfingstlichen Weingartener Alleluia-Reihe, wie von Scalon (1979, p. 148) bemerkt, liegt nicht vor: In verschiedenen Repertoireschichten erlaubt eine tragfähige Zuweisung an einen be- Weingarten verwendete man die von Heinzer (1992, p. 332) für den Hirsauer Reformkreis rekonstruierte Versfolge (so etwa stimmten Herstellungs- oder Bestimmungsort. Mit diesem Ansatz werde ich mich im in D-FUl Aa 6 und D-FUl Aa 32, zwei Gradualien des frühen zwölften und dreizehnten Jahrhunderts aus Weingarten), von der I-UDba Ms. 78 abweicht. Die in der Udineser Handschrift überlieferte Alleluia-Reihe deckt sich am ehesten mit Folgenden einem Ausschnitt der für Moggio zu sichernden musikalisch-liturgischen derjenigen von Sankt Peter in Salzburg (A-Ssp a IX 11, A-Wn Cod. Ser.n. 2700), allerdings werden die Verse für die Sonntage 17-20 in anderer Reihenfolge eingesetzt. Anders als von Heinzer (1997, p. 160*) vermutet, ist der für Melk charakteristische Überlieferung widmen. Heilige Koloman nicht im Nobis quoque peccatoribus des Canon missae genannt, dort (fol. 51) wird am Ende der Reihe der Jungfrauen die Heilige Columba angeführt. Koloman wird lediglich im Kalendarium (fol. 3v) berücksichtigt (neben Ausgangspunkt für die Frage nach den Spezifika der liturgischen Musikpraxis des weiteren in Österreich und Süddeutschland verehrten Heiligen wie beispielsweise Rupert mit Translationsfest, fols. 2, 3v). Die Zusammensetzung des Tropars von I-UDba Ms. 78 (fols. 123v-125) steht, wie Andreas Haug gezeigt hat (2008, pp. 157- friulanischen Klosters bildet die Untersuchung des eingangs zitierten Normtextes der 166), in der durch Quellen aus Freising, Moosburg und Innichen dokumentierten Tradition. Möglicherweise lässt sich die Provenienz der Handschrift, die bislang nur grob dem süddeutsch-österreichischen Raum zugewiesen werden kann, über Abtei (GB-Ob MS. Canon. Liturg. 325). Die gegen Ende des zwölften Jahrhunderts das im Sakramentar erwähnte Stephanus-Patrozinium (fol. 100) weiter eingrenzen. Auf eine frühe Präsenz des Codex in Moggio deuten die Nachträge im Kalendarium sowie ein Besitzeintrag des zwölften/dreizehnten Jahrhunderts hin, der in das friulanische Résiatal weist («Bernhardus dei gratia plebanus de resia», fol. 123v). 11. «A noi sembra di aver dimostrato che anche Moggio […] deve la sua origine al monachesimo riformato di Hirsau probabilmente attraverso la mediazione di S. Emmerano di Ratisbona» (Scalon 1979, p. 32). 9. Die illuminierten Handschriften aus Moggio im Bestand der Udineser Bibliothek überblickt Bergamini (1994). 12. Zur Hirsauer Tropen- und Sequenzenpraxis vgl. Haug (1994) und Kruckenberg (1999). Zum Hymnar der Reform­ 10. Vgl. von kunsthistorischer Seite ferner Pächt und Alexander (1966, pp. 5-9). bewegung vgl. Heinzer (2004).

312 313 CANTUS PLANUS 2014 ZÜHLKE — HIRSAU IN NORDITALIEN entstandene Handschrift diente Heinzer als Argumentationsfolie für die Identifikation und Weingarten hat ergeben, dass der friulanische Normtext an mehreren Stellen des Hirsauer Liber ordinarius in dem bislang als Rheinauer Liber ordinarius bekann- gegen die anderen Textzeugen steht (und zwar nicht nur mit Blick auf die oben ge- ten und edierten Text (Hänggi 1957; Heinzer 1992). Waren für Heinzer in erster Linie nannten aquileischen Lokalakzente). Die Unterschiede beziehen sich erstens auf die die Übereinstimmungen zwischen dem Normtext aus Moggio und den Libri ordinarii berücksichtigten Heiligenfeste, zweitens auf die Ausführlichkeit der Angaben zu den aus den Hirsauer Tochterklöstern Rheinau, Weingarten und Zwiefalten entscheidend, einzelnen Festen und drittens auf die Auswahl der Gesänge. Für die drei genannten werde ich mich im Folgenden vor allem auf die Unterschiede zwischen diesen konzen- Bereiche werden im Folgenden einige Beispiele angeführt. trieren. Denn gerade an denjenigen Stellen, an denen der Textzeuge aus dem friulani- schen Kloster von allen anderen Textzeugen des Hirsauer Liber ordinarius abweicht, In die Angaben zur zweiten Vesper am Fest der Kreuzauffindung (3. Mai) integriert werden die Besonderheiten der musikalisch-liturgischen Tradition Moggios greifbar. der Liber ordinarius aus Moggio eine Antiphon für den Heiligen Florian; für dessen Mit Hilfe des auf diese Weise konturierten Profils lässt sich in einem zweiten Schritt Oratio und Messe wird auf Tiburtius und Valerianus sowie Vitalis verwiesen (fol. 98). herausarbeiten, welche musikalisch-liturgischen Handschriften eindeutig für einen An Nativitas Mariae werden in Moggio neben Antiphonen für den Heiligen Adrian Gebrauch in Moggio hergestellt worden sind, und welche Codices zunächst für andere (wie in Rheinau und Weingarten) auch zwei Antiphonen für Corbinian gesungen. Die Hirsauer Klöster bestimmt waren und erst sekundäre Verwendung in Moggio fanden. vorgesehene Oratio soll dem Formular für den Heiligen Urban entnommen werden (fols. 125v, 126v). Weder Florian noch Corbinian sind in den Textzeugen aus Rheinau Die Anfertigung des «Ordo secundum Hirsiacenses» (GB-Ob MS. Canon. Liturg. 325) und Weingarten berücksichtigt; ihre Verehrung beschränkt sich weitgehend auf Süd- für die friulanische Abtei ist durch die Erweiterung des Festbestandes um für die deutschland und Österreich. Interessant ist die Verzahnung der Feste des Heiligen Diözese Aquileia typische Heiligenfeste gesichert: Berücksichtigt sind die Patrone von Hilarion und der Elftausend Jungfrauen in Moggio, die beide am 21. Oktober gefeiert Aquileia, Hermagoras und Fortunatus (fol. 116). Die am 12. Juli verehrten Heiligen werden: Während in den Libri ordinarii aus Rheinau und Weingarten die Jungfrauen werden überdies «infra octavam» gefeiert (fol. 116v). Helarius und Tatianus unter den lediglich mit zwei Antiphonen (in der ersten Vesper und in den Laudes) bedacht sind Heiligenfesten im März (fol. 49v) verleihen der Handschrift einen weiteren Lokalak- (Hänggi 1957, pp. 215-216), dreht der Liber ordinarius aus Moggio die Wertigkeit der zent. Die Verehrung des Heiligen Gallus «infra octavam» (fol. 132) trägt schließlich beiden Feste um und stattet die Jungfrauen mit weiteren Gesängen aus (fol. 133). Hier dem Gallus-Patrozinium von Moggio Rechnung. Unklar ist nach wie vor, wo genau ist Hilarion mit nur zwei Antiphonen präsent. Zudem wird in der Messe der Bezug die Handschrift entstanden ist. Wie im Falle der übrigen aus Moggio überlieferten auf die Jungfrauen in den Fokus gestellt (in Rheinau und Weingarten werden dagegen Codices des zwölften und dreizehnten Jahrhunderts deuten Textschrift, Buchschmuck die Gesänge vom Fest des Paulus Heremita für Hilarion verwendet; Hänggi 1957, p. und musikalische Notation in den Raum nördlich der Alpen; Otto Pächt und Jonathan 216). Die größere Bedeutung, die den Elftausend Jungfrauen in Moggio zukommt, Alexander (1966, p. 6) lokalisieren die Handschrift anhand der Illuminationen in das wird nicht zuletzt in der Rezeption neuer Gesänge greifbar, die im Liber ordinarius südliche Österreich. Dieser Bezug wird auch durch die Nachträge am Beginn her- vermutlich im späten vierzehnten Jahrhundert über Rasur eingetragen wurden, da- gestellt: In der auf fol. 3 eingetragenen Verbrüderungsnotiz ist eine monastische Ge- runter das Responsorium O sanctarum milia und die Antiphon O felix Germania aus meinschaft «aput Anesum» angesprochen, die sich nach Rudolf Flotzinger (1991, p. den beiden bekannten Reimoffizien für Ursula A( h 5-88, Ah 28-90). Auf diese verweist 42; 2002, pp. 263-265) auf die an der Enns gelegenen Klöster Admont oder Garsten offensichtlich auch die Rubrik «Cetera omnia sicut in antiphonario de nobis virginibus beziehen kann. Das Mittelhochdeutsch der auf fol. 12v aufgezeichneten Fassung der inuenitur Istoria propria» (fol. 133). Der ebenfalls später ergänzte Hymnus Virginale sogenannten «Mariensequenz aus Muri» weist Barbara Gutfleisch (1990) zufolge praeconium (Ah 43-506) ist in erster Linie durch Quellen des vierzehnten und fünf- in das ostoberdeutsche Sprachgebiet.13 Über den Vergleich mit weiteren Oxforder zehnten Jahrhunderts aus Cividale überliefert. Für Moggio wird er überdies durch das Handschriften aus dem ehemaligen Bestand der Klosterbibliothek in Moggio brachte Hymnar I-UDba Ms. 80 aus dem vierzehnten Jahrhundert belegt. Flotzinger (1991, pp. 38-66; 2002) den Liber ordinarius der friulanischen Abtei mit dem Hirsauer Tochterkloster Münsterschwarzach am Main bei Würzburg in Verbin- Unterschiede zwischen dem Liber ordinarius aus Moggio und den Normtexten aus dung. Diese Annahme wird im Folgenden noch einmal aufzugreifen sein. Für die Fra- Rheinau und Weingarten bezogen auf die Ausführlichkeit der Angaben zu einzel- ge nach der hochmittelalterlichen liturgischen Gesangspraxis Moggios ist zunächst nen Festen finden sich etwa bei Gervasius und Protasius: Moggio sieht für die erste Viri sancti ao Sanctorum meritis nur die eindeutige Bestimmung für das Kloster im Friaul entscheidend. Eine voll- Vesper das Responsorium (C 7906), den Hymnus h Sanctum est verum ao ständige Gegenlesung des Liber ordinarius aus Moggio mit den in der Edition Anton (A 50-153) und die Magnificat-Antiphon (C 4768) vor (fol. v Haec est vera Hänggis (1957) zugänglichen Textzeugen des Hirsauer Liber ordinarius aus Rheinau 110 ). In Rheinau und Weingarten ist nur die Magnificat-Antiphon fraternitas (Cao 3003) angegeben. Für die Nocturnen folgt in Moggio der Verweis 13. Zur Melodie der Sequenz vgl. Flotzinger (1990). auf das Commune für Märtyrer; Rheinau und Weingarten verzeichnen dagegen

314 315 CANTUS PLANUS 2014 ZÜHLKE — HIRSAU IN NORDITALIEN drei Responsorien (Hänggi 1957, p. 179). Die Angaben für die Laudes stimmen in dort fols. 25, 49, 64v, 66v); Florian und Corbinian im Übrigen in identischer Weise wie allen Textzeugen überein; sie differieren jedoch erneut an der zweiten Vesper: Hier in Moggio.14 trägt Moggio die Antiphonen Virgam virtutis (Cao 5442) und Fulgebunt justi (Cao 2907) ein, während Rheinau und Weingarten keine Gesänge vermerken. In gleicher Ich möchte vorerst in Moggio bleiben und den Blick auf zwei Breviere aus dem Be- Weise unterscheiden sich die Textzeugen am Fest des Heiligen Lambertus: Moggio stand der ehemaligen Klosterbibliothek lenken. Die Handschrift GB-Ob MS. Canon. verweist für die Nocturnen auf das Commune eines Bischof-Märtyrers (fol. 128); Liturg. 346, ein Brevier aus dem ersten Viertel des dreizehnten Jahrhunderts, konnte Rheinau und Weingarten verzeichnen drei Responsorien (Hänggi 1957, p. 207). Hier bislang aufgrund der Formulare für Hermagoras und Fortunatus (fols. 244, 245r-v) fehlen Angaben zur zweiten Vesper; Moggio sieht dagegen die Antiphonen Jucundus sowie der Oktavfeier für den Heiligen Gallus (fol. 276v) mit Moggio in Verbindung homo (mit dem Vermerk «cum reliquis», Cao 3510) und O quam venerandus (Cao gebracht werden (Flotzinger 1991, p. 49). Pächt und Alexander (1966, p. 9) vermu- 4071) vor. In der Auswahl der Gesänge besteht eine auffällige Differenz zwischen ten wie für GB-Ob 325 eine Entstehung im südlichen Österreich. Ein Vergleich mit Moggio und den übrigen Textzeugen am Fest des Heiligen Benedikt: Während die der im Liber ordinarius aus Moggio dokumentierten liturgischen Praxis (auch hier Libri ordinarii aus Rheinau und Weingarten hier auf das Prosaoffizium Fuit vir vitae lag der Fokus insbesondere auf den von den übrigen Textzeugen des Hirsauer Liber venerabilis zurückgreifen (Hänggi 1957, pp. 90-91), sieht der friulanische Normtext ordinarius abweichenden Stellen) ergab eine vollständige Übereinstimmung. GB-Ob das ReimoffiziumFlorem mundi periturum (Ah 25-52) vor (fols. 49v-50). Einzel- 346 berücksichtigt alle im Normtext verzeichneten Heiligenfeste (also auch Florian, ne Gesänge differieren etwa am Fest Assumptio Mariae oder am Tag des Heiligen Corbinian sowie Hilarion und die Elftausend Jungfrauen in der oben beschriebenen Remigius: Rheinau und Weingarten vermerken für Assumptio das relativ weit ver- Rangfolge) und deckt sich mit diesem exakt in der Ausführlichkeit der Angaben zu breitete Invitatorium In honore beatissimae Mariae (Cao 1086, Hänggi 1957, p. 198). den einzelnen Festen. Ferner speist sich der Gesangsbestand des Breviers aus der Moggio gibt hingegen das durch die Cantus-Datenbank lediglich in einem Brevier im Liber ordinarius getroffenen Auswahl von Gesängen, darunter das eben als Bei- des späteren zwölften Jahrhunderts aus dem hirsauisch geprägten Kremsmünster spiel angeführte Invitatorium Virginis aeterne celebrantes (fol. 254v) für Assumptio (A-Lls Cod. 290) und im Zwiefaltener Antiphonar D-KA Aug. LX belegte Invitato- und der Remigius-Vers (fol. 273v). Das Brevier wurde demnach mit Sicherheit vor rium Virginis aeterne celebrantes (Cid 100295) an. Für das Responsorium Iste est dem Hintergrund der im Normtext festgehaltenen liturgischen Tradition angefertigt, de sublimibus (Cao 6998), das in der dritten Nocturn für Remigius gesungen wird, wenn nicht sogar mit diesem als Vorlage. Es ist anzunehmen, dass beide vom glei- kennt der Liber ordinarius aus Moggio einen anderen Vers als die Normtexte aus chen Entstehungsort stammen. Etwas anders verhält es sich mit dem Brevier GB- Rheinau und Weingarten (Hänggi 1957, p. 211): Bezeichnender Weise handelt es sich Ob MS. Canon. Liturg. 297. Diese Handschrift zählt zu den ältesten Codices Hirsauer bei dem Vers Iste est enim Remigius cujus um den einzigen Gesang, der im Codex aus Prägung aus dem Bestand der ehemaligen Klosterbibliothek in Moggio, ihre Tabelle Moggio ausgeschrieben und vollständig neumiert ist (fol. 130). der Jahreskennzeichen beginnt 1154. Sie ist also älter als der aus Moggio überlieferte Liber ordinarius. Als Entstehungsort wird in der Literatur das Hirsauer Tochterkloster Für die Frage nach den Traditionszusammenhängen der in Moggio verwendeten hoch- Münsterschwarzach am Main angegeben (Berger 1963; Flotzinger 1991, pp. 60-62). mittelalterlichen Liturgie, wie sie der Liber ordinarius der Abtei dokumentiert, und Darauf deuten das Kalendarium der Handschrift, das für die Diözese Würzburg ty- damit letztlich für die Frage nach dem konkreten Entstehungsort des Normtextes wäre pische Heiligenfeste vermerkt, darunter die «Depositio sancti Burchardi episcopi in im Folgenden ein Vergleich mit der liturgischen Musikpraxis verschiedener Hirsauer Wirziburg» am 2. Februar (fol. 3v) sowie dessen Hauptfest am 14. Oktober (fol. 7v), Klöster im süddeutsch-österreichischen Raum zu unternehmen. Als besonders trag- ferner Kilian mit Gefährten am 8. Juli und die fränkische Heilige Reginswindis am fähig dürften sich hierfür die Differenzen zwischen dem Liber ordinarius aus Moggio 15. Juli (fol. 6). Ein weiteres Indiz für die Würzburger Provenienz ist die besondere und den übrigen Textzeugen des Hirsauer Liber ordinarius, wie oben beispielhaft zu- Hervorhebung der Klosterpatronin von Münsterschwarzach, der Heiligen Felicitas, in sammengestellt, erweisen. Dass sich die Probebohrungen wohl auf diesen geographi- der Handschrift (fols. 6, 258v-260, 319, 327). Als Hinweis auf die frühe Verwendung schen Raum beschränken lassen, wird neben der kunstgeschichtlichen Einordnung des Codex in Moggio führt Flotzinger (1991, p. 61) den Nekrolog-Eintrag eines «Al- jetzt auch durch den inhaltlichen Befund nahe gelegt: Verwiesen sei hier nochmals auf bertus presbyter et monachus sancti Galli» am 15. August an, der sich in weiteren die Berücksichtigung der Heiligen Florian und Corbinian; zu ergänzen wäre überdies der Patron der Salzburger Diözese, der Heilige Rupert, der im Vorspann über die ver- 14. Zu den Spezifika der liturgischen Musikpraxis des österreichischen Subzentrums der Hirsauer Reform vgl. schiedenen Feierlichkeitsgrade der Feste im friulanischen Normtext genannt ist (fol. Zühlke (2012; forthcoming). Für zwei der aus Moggio überlieferten Messhandschriften (GB-Ob MS. Canon. Liturg. 340, um 1210 und I-UDba Ms. 75, um 1200) ließ sich eine Anfertigung in Admont für Moggio wahrscheinlich machen 13). Die drei Heiligen sind auch in einem späten Textzeugen des Hirsauer Liber ordi- (Flotzinger 1991, pp. 49-66; Zühlke 2012, pp. 443-445, 450-451). Inwieweit auch der Liber ordinarius des friulanischen narius aus Admont berücksichtigt (A-A Cod. 790 aus dem fünfzehnten Jahrhundert; Klosters nach Admont verortet werden kann, bedarf der weiteren Erforschung, insbesondere einer systematischen Auswertung des Admonter Normtextes aus dem fünfzehnten Jahrhundert.

316 317 CANTUS PLANUS 2014 ZÜHLKE — HIRSAU IN NORDITALIEN

Kalendarien aus Moggio findet.15 Flotzingers These eines Auftragswerkes für das friulanische Kloster (ibidem; 2002, p. 265) scheint vor dem Hintergrund der eben für Moggio konturierten musikalisch-liturgischen Praxis jedoch nur schwer haltbar (und dies nicht nur aufgrund der völligen Absenz von für Moggio typischen Heiligenfesten): GB-Ob 297 berücksichtigt sehr viel weniger Feste als der Liber ordinarius und das jüngere Brevier, beispielsweise fehlen Hilarion und die Elftausend Jungfrauen, und die Angaben zu einzelnen Festen weichen vom Normtext ab. Bislang nicht in den Blick genommen hat man die zahlreichen Nachträge an den Seitenrändern der Handschrift, Abbildung 1. GB-Ob MS. Canon. Liturg. 297, fol. 241 die den Fest-, Gesangs- und Textbestand insbesondere ab den Heiligenfesten im Au- gust erweitern (siehe Abbildungen 1-3). Die Funktion der Nachträge ist offensichtlich: Für verschiedene Abschnitte des Kirchenjahres wurde der Codex an die im Liber or- dinarius und in GB-Ob 346 dokumentierte Praxis adaptiert. Nachgetragen sind etwa Corbinian (fols. 241, 243, und zwar in der gleichen Weise wie im Liber ordinarius aus Moggio berücksichtigt), Angaben zu Lambertus (fol. 244, darunter wie im Norm- text Gesänge für die zweite Vesper) oder Gesänge und Texte für Remigius (fol. 248v, darunter auch der ungewöhnliche Vers, wie in GB-Ob 325 in vollständiger Länge).16 Das Invitatorium für Assumptio Mariae wurde über Rasur im Haupttext ergänzt (fol. 233). Folgendes Szenario ist denkbar: Das Brevier GB-Ob 297 wurde um 1154 für eine Verwendung in Münsterschwarzach angefertigt und gelangte in der Folge (wann und auf welchem Weg auch immer)17 nach Moggio. Dort wurde es vor dem Hintergrund der im Liber ordinarius und in GB-Ob 346 kodifizierten Tradition überarbeitet (wie die Textschrift der Nachträge erkennen lässt, vermutlich nach der Entstehung des Normtextes und des jüngeren Breviers). Dass alle drei Handschriften aus Münster- Abbildung 2. GB-Ob MS. Canon. Liturg. 297, fol. 243 schwarzach nach Moggio gelangt sind (Flotzinger 1991, pp. 49, 61-62) oder mit Bezug auf den liturgischen Usus der unterfränkischen Abtei angefertigt wurden (Flotzinger 2002, pp. 260-261), scheint aufgrund des inhaltlichen Befundes wenig naheliegend.

Am Beispiel der Oxforder Codices zeigt sich einmal mehr die je spezifische Färbung der liturgischen Praxis verschiedener Hirsauer Tochterklöster, die im Falle eines Transfers von Handschriften sogar deren Überarbeitung herausfordern konnte. Der gemeinsame überregionale Traditionszusammenhang blieb gleichwohl stets erkenn- bar, wenn auch nicht immer so sprechend wie am Beginn des friulanischen Norm- textes als «Ordo secundum Hirsiacenses».

15. Darunter auch das Kalendarium der in Anmerkung 8 diskutierten Handschrift I-UDba Ms. 78.

16. Es handelt sich hierbei offensichtlich um ein Kontrafaktum des für Johannes Baptista verwendeten Verses Iste est enim Joannes cujus (Cao 6998b). An weiteren Belegen konnte ich bislang zusammentragen: A-A Cod. 790 (Admont, fünfzehntes Jahrhundert), fol. 67; A-Lls Cod. 290 (Kremsmünster, späteres zwölftes Jahrhundert), fol. 357; A-Wn Fragm. 960 (Mondsee, zwölftes Jahrhundert), fol. 5b; CZ-Bu R 387 (Hradisko bei Olomouc/Rajhrad bei Brno, zwölftes Jahrhundert), fol. 82; CZ-Pu VI E 13 (Sankt Georg in Prag, zwölftes Jahrhundert), p. 434; weitere Quellen aus dem Prager Benediktinerinnenkloster: CZ-Pu XXIII D 156 (dreizehntes Jahrhundert), fol. 124v; CZ-Pu XIII C 4 (vierzehntes Jahrhundert), fol. 11v; CZ-Pu XIV C 20 (vierzehntes Jahrhundert), fol. 10v; D-KA Aug. LX (Zwiefalten, zweite Hälfte des zwölften Jahrhunderts), fol. 158v (Nachtrag); D-Mbs Clm 23037 (Prüfening, um 1140), fol. 260. Inwieweit die Verbreitung des Verses tatsächlich an ein Subnetzwerk des Hirsauer Kreises geknüpft ist, wie der momentane Quellenbefund nahelegt, wäre zu prüfen.

17. «Il manoscritto 297 potrebbe essere giunto a Moggio attraverso la mediazione di Millstatt, oppure direttamente da Münsterschwarzach» (Flotzinger 2002, p. 271). Abbildung 3. GB-Ob MS. Canon. Liturg. 297, fol. 248v

318 319 CANTUS PLANUS 2014 ZÜHLKE — HIRSAU IN NORDITALIEN

Bibliography Heinzer Felix (1992), ‘Der Hirsauer «Liber ordinarius»’, Revue Bénédictine 102, pp. 309-347; reprinted in Heinzer 2008, pp. 185-223. Baum Wilhelm (2002), Rosazzo in Ulrich Faust - Waltraud Krassnig eds., Die benediktinischen Mönchs- und Nonnenklöster in Österreich und Südtirol, Sankt Ottilien: Eos-Verlag (Germania Benedictina 3/3), – (1997), ‘Rezension zu Christine Glaßner und Alois Haidinger «Die Anfänge der Melker Bibliothek. pp. 152-182. Neue Erkenntnisse zu Handschriften und Fragmenten aus der Zeit vor 1200 […]» Melk 1996’, Scriptorium 51, pp. 159*-161* (B 497). Bergamini Giuseppe (1994), Codici miniati dell’Abbazia di Moggio in Menis 1994b, pp. 85-103. – (2004), Liturgischer Hymnus und monastische Reform. Zur Rekonstruktion des Hirsauer Hymnars in Berger Placidus (1963), Zum Kalendarium eines Schwarzacher Breviers von 1154 in Adelhard Kaspar - Andreas Haug - David Hiley - Karlheinz Schlager eds., Der lateinische Hymnus im Mittelalter. Über­ Alfred Wendehorst eds., Studia Suarzacensia. Beiträge zur Geschichte der Abtei Münsterschwarzach­ lieferung. Ästhetik. Ausstrahlung, Kassel: Bärenreiter (Mmmae, Subsidia 4), pp. 23-52; reprinted in anläßlich des 50. Jahrestages ihrer Wiederbesiedelung, Münsterschwarzach: Vier-Türme-Verlag Heinzer 2008, pp. 224-256. (Würzburger Diözesangeschichtsblätter 25), pp. 113‑124. – (2008), Klosterreform­ und mittel­alterliche Buchkultur­ im deutschen Südwesten, Leiden-Boston: Brill. Bischoff Bernhard (1980), Die südostdeutschen Schreibschulen und Bibliotheken in der Karolingerzeit, 2 vols., Wiesbaden: Harrassowitz, vol. 2: Die vorwiegend österreichischen Diözesen. Huglo Michel (1976), Liturgia e musica sacra aquileiese in Gianfranco Folena ed., Storia della cultura veneta, 6 vols., Vicenza: Neri Pozza, vol. 1: Dalle origini al trecento, 1, pp. 312-325. Camilot-Oswald Raffaella (1997), Die liturgischen Musikhandschriften aus dem mittelalterlichen Patriarchat Aquileia, Teilband 1: Einleitung, Handschriftenbeschreibungen, Kassel: Bärenreiter (Mmmae, Subsidia Jakobs Hermann (1961), Die Hirsauer. Ihre Ausbreitung und Rechtsstellung im Zeitalter des Investiturstreites­ , 2/1). Köln-Graz: Böhlau (Kölner Historische Abhandlungen 4).

Casarsa Mario (1968), I codici liturgici dell’Abbazia di Moggio, Udine: Arti Grafiche Friulane. Kruckenberg Lori (1999), ‘Zur Rekonstruktion des Hirsauer Sequentiars’, Revue Bénédictine 109, pp. 186‑207. Flotzinger Rudolf (1990), ‘Zur Melodie der sog. «Mariensequenz aus Muri» in Can. lit. 325 der Bodleian Library in Oxford’, Zeitschrift für deutsches Altertum und deutsche Literatur 119, pp. 75-82. Menis Gian Carlo (1994a), Le origini dell’Abbazia di Moggio in Menis 1994b, pp. 7-15.

– (1991), Choralhandschriften österreichischer Provenienz in der Bodleian Library/Oxford, Wien: Menis Gian Carlo ed. (1994b), Le origini dell’Abbazia di Moggio e i suoi rapporti con l’Abbazia svizzera di Verlag der Österreichischen Akademie der Wissenschaften (Veröffentlichungen der Kommission für San Gallo. Atti del convegno internazionale Moggio 5 dicembre 1992, Udine: Arti Grafiche Friulane Musikforschung 26). (Pubblicazioni della Deputazione di Storia Patria per il Friuli 21).

– (2002), Analisi musicologica di manoscritti medievali: i testimoni mosacensi in Cesare Scalon ed., Il Ochsenbein Peter (1993), ‘Moggio di San Gallo. Ein bislang unbekanntes Kloster der Hirsauer Reform’, monachesimo benedettino in Friuli in età patriarcale. Atti del Convegno Internazionale di Studi Udine - Studien und Mitteilungen zur Geschichte des Benediktinerordens und seiner Zweige 104, pp. 315‑327; It. Rosazzo, 18-20 Novembre 1999, Udine: Forum (Studi per la storia della chiesa in Friuli 3), pp. 257-274. transl. Tracce della liturgia sangallese nei codici moggesi di Oxford e Udine in Menis 1994b, pp. 65‑78; reprinted in Ochsenbein Peter (2000), Cultura Sangallensis. Gesammelte Aufsätze zu seinem 60. Foligno Cesare (1913), ‘Di alcuni codici liturgici di provenienza friulana nella Biblioteca Bodleiana di Geburtstag, Sankt Gallen: Verlag am Klosterhof (Monasterium Sancti Galli 1), pp. 49-63. Oxford’, Memorie storiche forogiuliesi 9, pp. 292-300. Pächt Otto - Alexander Jonathan J. G. (1966), Illuminated manuscripts in the Bodleian Library Oxford, Gutfleisch Barbara (1990), ‘Eine ostoberdeutsche Handschrift der «Mariensequenz aus Muri»’, Zeitschrift 3 vols., Oxford: Clarendon Press, vol. 1: German, Dutch, Flemish, French and Spanish Schools. für deutsches Altertum und deutsche Literatur 119, pp. 61-75. Scalon Cesare (1979), La Biblioteca Arcivescovile di Udine, Padova: Antenore (Medioevo e Umanesimo 37). Hänggi Anton ed. (1957), Der Rheinauer Liber Ordinarius (Zürich Rh 80, Anfang 12. Jh.), Freiburg, CH: Universitätsverlag (Spicilegium Friburgense 1). – (1984), Fonti e ricerche per la storia del monastero benedettino di S. Maria di Aquileia in Giuseppe Fornasir ed., Il Friuli dagli Ottoni agli Hohenstaufen. Atti del convegno internazionale di studio, Härtel Reinhard (1985), Die älteren Urkunden des Klosters Moggio (bis 1250), Wien: Verlag der Udine 4-8 dicembre 1983, Udine: Arti Grafiche Friulane (Deputazione di Storia Patria per il Friuli), Österreichischen Akademie der Wissenschaften (Publikationen des Historischen Instituts beim pp. 53‑189. Österreichischen Kulturinstitut in Rom 2/6/1). – (1987), Über einige Handschriftenfragmente­ aus Moggio in Reinhard Härtel ed., Geschichte und ihre – (1994), Le fonti diplomatiche e la fondazione dell’Abbazia di Moggio in Menis 1994b, pp. 17-44. Quellen. Festschrift für Friedrich Hausmann zum 70. Geburtstag, Graz: Akademische Druck- und Verlagsanstalt,­ pp. 623-628. – (2001), Moggio in Ulrich Faust - Waltraud Krassnig eds., Die benediktinischen Mönchs- und Nonnenklöster in Österreich und Südtirol, Sankt Ottilien: Eos-Verlag (Germania Benedictina 3/2), Vogler Werner (1994), Ulrico di Eppenstein e l’Abbazia di San Gallo di Moggio in Menis 1994b, pp. 45-64. pp. 832-862. Zühlke Hanna (2012), Die musikalisch-liturgischen Quellen des Hochmittel­alters aus dem österreichischen Haug Andreas (1994), ‘Ein «Hirsauer» Tropus’, Revue Bénédictine 104, pp. 328-345. Benediktinerstift Admont in Robert Klugseder ed., International Musicological Society Study Group Cantus Planus. Papers read at the 16th meeting Vienna, Austria 2011, Wien: Hollinek, pp. 438-451. – (2008), Tropen im südostdeutschen und im nordostitalienischen Raum: Untersuchungen zu ihren Über­ lieferungs­wegen in Wulf Arlt - Giulio Cattin eds., Itinerari e stratificazioni dei tropi. San Marco, – (forthcoming), Zur liturgischen Musikpraxis des Benediktinerklosters Admont im Mittelalter in Irene l’Italia settentrionale e le regioni transalpine, Venezia: Fondazione Levi, pp. 137-175. Holzer - Maike Smit-Schilling eds., Rezeption. Produktion. Tradition. Liturgische Musik in der mittel­alter­lichen Kirchenprovinz Salzburg, Würzburg: Königshausen & Neumann.

320 321 Nausica Morandi

The office for Saint Cecilia in musico-liturgical manuscripts from the Pontificia Biblioteca Antoniana of Padua

This work examines the diffusion of the office for Saint Cecilia in the particular liturgical and musical tradition of the basilica of Saint Anthony of Padua. The chants of the office in the antiphonaries and breviaries preserved in the Biblioteca Antoniana were compared to another similar corpus, mainly of northern-Italian origin. A detailed system of synchronic and diachronic comparison and synoptic tables was used to carry out this analysis. The study was then expanded to incorporate hagiographical Latin manuscripts on literary subjects kept at the same library, including the Legenda aurea by Iacopo de Varagine, and two books from the Liber Epilogorum by Bartolomeo da Trento, in which the passio of Cecilia is presented as an exemplum sanctitatis.

The Pontificia Biblioteca Antoniana of Padua (I-Pca) The Pontificia Biblioteca Antoniana is probably the only Franciscan library in Italy today that has survived the centuries intact, notwithstanding sackings, dispersions and confiscations.1 The earliest records of the conventual library date from 1234, with reference in several documents to a room ove se leze (for reading) that was located in a small courtyard (Re 1929, pp. 216-217). The friars who were responsible for acquiring the first manuscripts and codices in that collection did not enjoy the benefits of patronage, but had to provide for themselves by purchasing or copying the documents they needed for study and teaching. In the Narbonne Constitutions, issued for the Franciscans in 1260, Saint Bonaventure made allowance for manuscript copying as a suitable occupation for friars and even for novices, whose studies were accordingly suspended during the novitiate year (Abate 1935, p. 23, rubric vi, col. 88). Saint Francis, inflexible in matters of religious poverty, allowed the use of books only in rare instances, and the Constitutiones Antiquae continued to forbid friars from receiving or accumulating money or other objects of value, with the sole exception of books and manuscripts (Abate 1935, pp. 8-16; Humphreys 1964, pp. 57-201; Maranesi 2000, pp. 23-155). For these reasons, the library’s collection was able to continue to grow even after Saint Anthony’s death, despite the historical vicissitudes that beset the convent and basilica in Padua. By the fifteenth century, the Biblioteca Antoniana had risen to a position of particular prestige owing to the achievements of Saint

1. Although also considered intact, the Library of Assisi is no longer situated in its original location (Luisetto 1999, pp. 62-66).

323 CANTUS PLANUS 2014 MORANDI — THE OFFICE FOR SAINT CECILIA

Anthony’s Studium.2 The guiding principles of the Biblioteca Antoniana are to be the office for Cecilia in the source from the Antoniana, indicating the liturgical act, sought in the intense pastoral and scientific activity of the Franciscan Order: the the type of chant, incipit, explicit, modality, and concordances with the other sources disciplines of theology, philosophy, history, biblical exegesis, and classical culture (listed in chronological order). The asterisk (*) denotes a textual piece without music. are all readily apparent in the volumes still conserved in the library today. The need The orthography has been normalized. For the abbreviations in all the tables see for a well-furnished library was also driven by the large number of friars coming appendix, p. 335. from all over Italy and Europe, attracted to the celebrated Archiginnasium of Padua and its Faculty of Theology. In turn, the library’s collection of medieval manuscripts Table 1. Office for Cecilia, Padua, Pontificia Biblioteca Antoniana, RR14 was enriched by the continual flow of documents of diverse provenience, brought to Act chant incipit explicit mode RR14 concordances the basilica by the visiting friars. 1vs amg Est secretum corpus meum 1 74 I-BGc M IV valeriane The office for Saint Cecilia in the music-liturgical manuscripts of the Biblioteca Antoniana 1vs ain Regem virginum venite adoremus 74v I-UDba 79, ; I-Fi Conv. sopp. dominum 560, I-Lc 601, I-BV 21, F-Pnm Of the thirteen parchment choral books and antiphonaries of various dimensions and Nouv. acq. lat. 1412, I-GO B, epochs kept in the Biblioteca Antoniana,3 only one, I- Pca S. Bach RR14 (henceforth I-ASs 1 RR14), fols. 73v-93, fourteenth century, Antiphonae a die 1 novembris usque ad 1nc 1ant Cecilia virgo coronas vocabat 8 75 I-UDba 79, I-SDF 4, ; I-Fi diem 30 eiusdem mensis attests to the office for Cecilia. The breviaries preserved in almachium Conv. sopp. 560, I-Lc 601, the Antoniana: I-Psa 88 Scaff. V, third quarter of the fifteenth century,Breviarium FaNn, F-Pnm Nouv. acq. lat. Franciscanum (small choral book) and I-Psa 104 Scaff. VI, mid-thirteenth century, 1412, I-GO B, I-ASs 1, I-BGc M IV Ordo breviarii et missalis fratrum minorum do not contain the passio of Cecilia at all, favouring the local cult of Saint Prosdocimus on November 7 instead. The other 1nc 2ant Expansis eriperet de 8 75v I-UDba 79, I-SDF 4, ; I-Fi manibus orabat inimicis Conv. sopp. 560, I-Lc 601, antiphonaries hold no entries for the month of November. FaNn, F-Pnm Nouv. acq. lat. 1412, I-GO B, I-ASs 1, I-BGc Analysis of the liturgy of the Hours M IV Comparative analysis was performed between the office for Cecilia contained in 1nc 3ant Cilicio Cecilia gemitibus 8 75v I-Udba 79, I-SDF 4, ; I-Fi antiphonary RR14 and other sources, mainly from northern Italy, selected as being membra exorabat Conv. sopp. 560, FaNn, typically representative of a particular liturgical tradition (Morandi 2006, pp. 71-151). F-Pnm Nouv. acq. lat. 1412, I-GO B, I-ASs 1, I-BGc M IV Because of the wide provenience of the manuscripts from the Biblioteca Antoniana, the analysis also covered a few sources from central and southern Italy.4 Table 1 shows 1nc vsc Specie tua et pulchritudine tua 76 I-Lc 601, FaNn, F-Pnm Nouv. pulchritudine acq. lat. 1412, I-GO B, I-ASs 1

1nc rsb Intende prospere procede et regna 76* 2. The founding of the Faculty of Theology at the University of Padua (1363), with the involvement of the different religious orders, gave an important impulse to the development of the Biblioteca Antoniana. In 1437, the conventual Studium of Saint procede Anthony acquired the status of Studium Generale, meaning that it was considered at the same level as the University. In 1630, after a request by the Venetian Government, Urban viii officially declared the Studium Generale a theological Faculty. 1nc 1rsp Cantantibus ut non confundar 8 76 I-UDba 79, I-SDF 4, ; I-Fi The Biblioteca Antoniana was opened to the public and became one of the most important consultation libraries in Padua organis Cecilia Conv. sopp. 560, I-Lc 601, (Luisetto 1999, p. 64; Abate 1950, pp. 101-104). FaNn, F-Pnm Nouv. acq. lat. 3. Records of the oldest choral books can be found in the inventories of the sacristy from 1396 and 1421. Choral books 1412, I-GO B, I-ASs 1, I-BGc were formerly preserved in the sacristy and then in the choir of the basilica for liturgical use. The oldest were destroyed M IV by fire in 1749 and new choral books dated 1760-1761 were commissioned just to replace them (Abate and Luisetto 1975, pp. 15-32). 1nc 2rsp O Beata Cecilia angelico 3 77 I-UDba 79, I-SDF 4, ; I-Fi 4. I-ASs 1: Asti, Biblioteca del Seminario, 1, antiphonary (Cistercian?), from Piedmont, fourteenth-fifteenth century; I-BGc que duos demoastrasti Conv. sopp. 560, I-Lc 601, M IV: Bergamo, Biblioteca Civica Angelo Mai, MIA IV, antiphonary, from Bergamo, fifteenth century; I-BV 21: Benevento, FaNn, F-Pnm Nouv. acq. lat. Biblioteca Capitolare, 21, antiphonary, from Benevento, twelfth-thirteenth century; Fonte Avellana, Archivio Eremo S. Croce breviary, from North Central Italy, eleventh-twelfth century (henceforth: Fa Nn); I-FI Conv. sopp. 560: Firenze, 1412, I-GO B, I-ASs 1, I-BGc Biblioteca Medicea Laurenziana, Conv. Soppr 560, antiphonary, from Tuscany-Vallombrosa, twelfth century; I-GO B: M IV Gorizia, Biblioteca Seminario Teologico Centrale, B, antiphonary, from Aquileia, thirteenth-fourteenth century; I-Lc 601: Lucca, Biblioteca capitolare, 601, antiphonary, from Pontetetto S. Maria, twelfth century; F-Pnm Nouv. acq. lat. 1412: Paris, 1nc 3rsp Virgo gloriosa divinis et 8 78v I-UDba 79, I-SDF 4, FaNn, Bibliothèque nationale de France, n.a.l. 1412, Cistercian antiphonary, from Morimondo, S. Maria, twelfth century; I-SDF semper oratione F-Pnm Nouv. acq. lat. 1412, 4: San Daniele del Friuli, Biblioteca Guarneriana, 4, breviary, from Aquileia, twelfth century; I-UDba 79: Udine, Biblioteca Arcivescovile, 79, breviary, from Ravenna-Pomposa, eleventh century. I-GO B, I-ASs 1, I-BGc M IV

324 325 CANTUS PLANUS 2014 MORANDI — THE OFFICE FOR SAINT CECILIA

Act chant incipit explicit mode RR14 concordances Act chant incipit explicit mode RR14 concordances

1nc 1vrs Biduanis ac domino quod 8 77v I-UDba 79, I-SDF 4, ; I-Fi 3nc rsb In tabernaculum 86v* triduanis timebat Conv. sopp. 560, I-Lc 601, FaNn, F-Pnm Nouv. acq. lat. 3nc 1rsp Beata Cecilia contemptorem 8 86v I-SDF 4, I-GO B, I-BGc M IV 1412, I-GO B, I-ASs 1, I-BGc dixit idolorum M IV 3nc 2rsp Cecilia me misit secreta que 4 88 I-SDF 4, I-GO B, I-BGc M IV 1nc 2vrs Cecilia me misit ostendatus michi 3 78v I-UDba 79, I-SDF 4, I-GO B, perferam ad vos I-BGc M IV 3nc 3rsp Dum aurora armis lucis 7 89 I-SDF 4, F-Pnm Nouv. acq. 1nc 3vrs Et secretum corpus meum 8 79v I-SDF 4, I-GO B, I-BGc M IV finem lat. 1412, I-GO B, I-ASs 1, Valeriane I-BGc M IV

2nc 1ant Domine Ihesu in Cecilia 8 80 I-SDF 4, F-Pnm Nouv. acq. 3nc 1vrs Sicut enim amor faciat esse 8 87v Christe seminasti lat. 1412, I-GO B, I-ASs 1, I-BGc M IV 3nc 2vrs Tunc Valerianus sanctus Urbanus 3 88v I-SDF 4, I-GO B, I-BGc M IV perrexit 2nc 2ant Beata Cecilia contemptorem 8 80v I-SDF 4, I-GO B, I-ASs 1, dixit idolorum I-BGc M IV 3nc 3vrs Cecilia valedicens exortans eos ait 7 89v I-SDF 4, F-Pnm Nouv. acq. fratribus lat. 1412, I-GO B, I-ASs 1 2nc 3ant Fiat Domine cor ut non confundar 2 81 I-SDF 4, I-GO B, I-ASs 1, meum I-BGc M IV laud amg Cantantibus ut non confundar 1 90 organis Cecilia 2nc vsc Adiuvabit eam 81v F-Pnm Nouv. acq. lat. 1412, deus I-ASs 1 laud 1ant Valerianus in orantes invenit 7 90v I-BV 21 cubiculo 2nc rsb Deus in medio 81v* eius laud 2ant Cecilia famula argumentosa 8 91 tua deservit 2nc 1rsp Cilitio Cecilia coronas vocabat 8 81v I-SDF 4, I-BGc M IV membra laud 3ant Benedico te pater latere meo 7 91

2nc 2rsp Ceciliam intra terrore corruptus 3 82v I-SDF 4, I-BGc M IV laud 4ant Triduanas a ecclesiam 8 91v I-BGc M IV cubiculum est domino consecrarem

2nc 3rsp Domine Ihesu argumentosa 8 84 I-SDF 4, I-BGc M IV laud vsc Diffusa est gratia 92 F-Pnm Nouv. acq. lat. 1412, Christe deservit I-ASs 1

2nc 1vrs Hec est virgo numero 8 82v I-BGc M IV laud rsb Propterea 92* prudentum laud abd Dum aurora armis lucis 1 92 I-UDba 79, I-SDF 4, ; I-Fi 2nc 2vrs Angelus domini habitaculo 3 83v I-SDF 4, I-GO B, I-BGc M IV finem Conv. sopp. 560, I-Lc 601, descendit Domine FaNn, I-BV 21, I-ASs 1, I-BGc M IV 2nc 3vrs Nam sponsum mansuetissimum 8 85 I-SDF 4, I-BGc M IV que destinavit 2vs amg Cantantibus 93* organis 3nc 1ant Credimus elegit famulam 1 85v I-SDF 4, I-GO B, I-BGc M IV Christum filium

3nc 2ant Nos scientes negare non 4 86 I-SDF 4, I-GO B, I-BGc M IV sanctum possumus Starting with the first nocturne, the source from the Antoniana, RR14, shows a high level of concordance with most of the other sources analyzed. The only exceptions are from 3nc 3ant Tunc Valerianus sanctum 8 86 I-SDF 4, I-GO B, I-BGc M IV perrexit Urbanum Morimondo (near Milan, F-Pnm Nouv. acq. lat. 1412) and San Daniele (I-SDF 4). In the responsorial verses, instead, concordance with RR14 is mainly limited to the sources 3nc vsc Elegit eas 86v F-Pnm Nouv. acq. lat. 1412, I-ASs 1 from Ravenna, Aquileia, Bergamo and Piedmont (I-UDba 79, I-GO B, I-SDF 4, I-BGc, M IV and I-ASs 1). In the second nocturne, concordances with RR14 are restricted to the

326 327 CANTUS PLANUS 2014 MORANDI — THE OFFICE FOR SAINT CECILIA sources from Bergamo and Aquileia, while those from Morimondo and Piedmont are for Cecilia from the Biblioteca Antoniana. Also worth mentioning are three cases where the only ones that share its structure for the versicle. Concordances with the sources the source from the Antoniana is concordant with only one other source because of from Bergamo and Aquileia are also persuasive in the third nocturne. In analyzing the the ordering of the chants. The placement of the antiphons Est secretum Valeriane (fol. nocturnes, particularly the second and the third, a strong connection emerges between 74) and Triduanas a domino (fol. 91v) in RR14 corresponds only to the source from RR14 and the sources from Aquileia (Gorizia and San Daniele) and from Bergamo. Bergamo, and of even greater note, the antiphon for the lauds Valerianus in cubiculo The dating of the sources and other elements arising from the analysis suggest that (fol. 90v) is consistent only with the Beneventan source. the earlier sources from Aquileia-San Daniele (twelfth century), and Aquileia-Gorizia (thirteenth-fourteenth century) likely influenced the later sources from the Antoniana The chants of the proprium (fourteenth century) and Bergamo (fifteenth century). The Benedictus antiphon for the The richly illuminated antiphonary RR14 has mainly been studied in terms of its lauds Dum aurora finemattests the concordance of all the examined sources. The office iconography, with only a few marginal studies focussing on its musical characteristics for Cecilia from the Antoniana also contains some peculiar chants, particularly the texts (Canova Mariani 2010, pp.380-400; Abate and Luisetto 1975, pp. 175-178). In the office of short responsories, of which there is no trace in any of the other sources examined. for Cecilia from the Biblioteca Antoniana, the antiphons do not conclude with the seculorum amen formula observed in the other sources, but present the incipit of a The short responsory Intende prospere procede (fol. 76) is a widely attested verse for psalm instead. Closer analysis of the antiphons reveals with even greater clarity the the Commune Virginum that is also used in the offices for Agnetis, Valeriae, Agathae, concordance between RR14 and the source from Bergamo: not only does the melodic Rictrudis, and in the Purificatio, Annuntiatio and Visitatio Mariae. The only two line match, but also that of the neumes. For example, in the widely adopted antiphon sources attesting this verse for Cecilia are two early French sources. The first is F-Pnm Dum aurora finem,RR14 contains more variants than the northern Italian sources, but Lat. 1085, an abbreviated antiphonary of the last quarter of the tenth century from the it matches the source from Bergamo, displaying almost identical melodies and neumes, Benedictine abbey of Saint Martial de Limoges (Collamore 2000, pp. 1-255; Grier 1995, and, notably, frequent changes of key. The source from the Antoniana also agrees with pp. 53-119). The numerous additions in the text and the margins by many different the group of thirteenth-century Franciscan sources, represented in example 1, p. 330, hands indicate a manuscript that was in use for a long period of time. The manuscript by the source from Fribourg, although the lectio of the Antoniana proves to be more shows every sign of being copied by a cantor for his own use, since the majority of the closely related to the later source from Bergamo. antiphons and responsories are abbreviated. The second source that testifies this chant for Cecilia is F-Pnm Lat. 12601, a breviary dating from between 1064 and 1095 probably In the office for Cecilia from the Antoniana, the antiphon Nos scientes sanctum is in the originated at Cluny and used in the monastery of Saint Taurin l’Echelle (Leroquais seventh mode, consistent only with the source from Bergamo (I-SDF 4, an adiastematic 1934, 3, pp. 226-228; Hourlier 1959, pp. 163-173; Lamothe-Constantine 1986, pp. 5-22; source, agrees only in the articulation of the neums). The same antiphon occurs in the Steiner 1993, pp. 175-204; Ferreira 1997, pp. 45-48). The same occurs for the Antoniana seventh mode in only fourteen of the seventy sources indexed by Cantus, and eight of short incipit Propterea (fol. 92), referred to in the aforementioned French sources as those fourteen are contained in the very same Franciscan sources from central Italy Propterea benedixit te, another verse used in the Commune Virginum. In Tabernaculum mentioned above that attest to the rhythmic office for Saint Anthony of Padua. The (fol. 86v) and Os in medio eius (fol. 81v) are found exclusively in the office for Cecilia Antoniana source also provides a short version of the long responsory Cantantibus from the source at the Antoniana. Sicut enim amor (fol. 87v) is attested to by more than organis Cecilia, while it lacks the short aside ut non confundar contained in all of thirty sources of different epochs and provenience.5 While identifying a persuasive the other sources examined. The verses Cecilia me misit (response: O Beata Caecilia) connection between the source from the Antoniana may seem daunting, important and Est secretum Veleriane (response: Virgo gloriosa semper) are attested only by the clues can be found in a group of four thirteenth-century Franciscan antiphonaries and twelfth-century sources from Aquileia and Ravenna and again by the later fifteenth- breviaries from central Italy that attest to the verse.6 All four sources contain records century source from Bergamo. The verse Hec est Virgo (response: Cilicio Cecilia of the rhythmic office for Saint Anthony of Padua, mainly added as later annotations. membra), attested by RR14, finds confirmation only in the source from Bergamo: Only fourteen surviving sources testify to the office for Saint Anthony, and all of them the melodies, with the exclusion of some minor variants, are substantially the same. have clear and direct connections with Padua. This makes the four antiphonaries and Extending the analysis to Cantus sources, the verse occurs in only ten of the more than breviaries likely candidates as sources for the verse Sicut enim amor used in the office eighty indexed sources attesting the office for Cecilia. Once again, of those ten, seven are some of the Franciscan sources containing the rhythmic office for Saint Anthony.7

5. http://cantusdatabase.org/

6. CH-Fco 2: Fribourg, Bibliothèque des Cordeliers, 2; I-Ad 5: Assisi, Cattedrale San Rufino, Archivio e Biblioteca, 5; 7. CH-Fco 2, twelfth century; D-Ma 12o Cmm 1; H-Bu lat. 121, Budapest, Egyetemi Könyvtár (University Library), lat. 121, US-Cn 24: Chicago, Newberry Library, 24: I-Nn vi.E.20: Napoli, Biblioteca nazionale Vittorio Emanuele iii, vi. E. 20. fourteenth century; and, from the thirteenth century: I-Ad 5; I-Nn vi.E.20; V-CVbav lat. 8737; US-Cn 24.

328 329 CANTUS PLANUS 2014 MORANDI — THE OFFICE FOR SAINT CECILIA

Example 1. Antiphon Dum aurora finem Example 2. Verse Hec est virgo 1. Fribourg, Bibliothèque des Cordeliers, 2 (CH-Fco 2) 1. Fribourg, Bibliothèque des Cordeliers, 2 (CH-Fco 2) 2. Padova, Biblioteca Antoniana, RR14 (I-Pca RR14) 2. Padova, Biblioteca Antoniana, RR14 (I-Pca RR14) 3. Bergamo, Biblioteca A. Mai, MIA IV (I-BGc Mia IV) 3. München Bayerische Staatsbibliothek, Clm 4305 (D-Mbs Clm 4305) 4. Bergamo, Biblioteca A. Mai, MIA IV (I-BGc Mia IV)

330 331 CANTUS PLANUS 2014 MORANDI — THE OFFICE FOR SAINT CECILIA

Two of the three remaining sources are Franciscan, while the third is a fifteenth- Table 2. Padua, Pontificia Biblioteca Antoniana, manuscripts on literary subjects containing episodes of the life and passio of Cecilia century antiphonary from the Benedictine monastery of Saints Ulrich and Affra in Augsburg (D-Mbs Clm 4305), the oldest complete office chant tradition of the so-called Date Author Title Fol./Col. Edition Melk Reform. Members of the reform movement at the monastery of Melk adopted the tradition of the Benedictine monastery in Subiaco. The Subiaco tradition cannot 81 fourteenth Iacobus Legendae sanctorum 247vb-250vb Legenda Aurea, century de Varagine De S. Cecilia;166 be defined as an old Italian, Benedictine tradition, such as that preserved in the Lucca antiphoner (I-Lc 601), but as one based on the usage of the papal court in the Lateran 98 fourteenth Iacobus Legenda aurea 236a-240a Legenda Aurea, century de Varagine De S. Cecilia; 149 church, which seems to have been instituted in the early thirteenth century (Schaff 1997) during the pontificate of the popes Innocentiii (1198-1216) and Honorius iii 101 fourteenth Iacobus Legenda aurea 198b-201b Legenda Aurea, century in. de Varagine De S. Cecilia; 168 (1216-1227). The question of how far the canons of Saint Fridianus in Lucca influenced this tradition remains to be answered. This train of events explains the similarity of 107 fourteenth Iacobus Legenda aurea 354b-359b Legenda Aurea, the Augsburg codices in the Melk tradition to early Franciscan codices. Much the same century, de Varagine De S. Cecilia; 173 first quarter pattern is to be observed in books of the Olivetan reform, beginning in the year 1313, which share the same order of chants and the rubrica romana. All of these reforms 219 fourteenth Iacobus Legenda aurea 331a-336a Legenda Aurea, century de Varagine (in the ms: Legenda De S. Cecilia; 165 are marked by an adaptation of the secular to the monastic cursus, an abbreviation first quarter sanctorum) of the office, a uniform practice of modal assignments and differentiae, and, finally, 470 thirteenth Batholomaeus Liber epilogorum 259vb-261a Liber epilogorum, a limitation of the ambitus (Nardini 2008, pp. 182-197). At first glance, the Melk century, Tridentinus in gestis sanctorum De S. Cecilia v. et antiphonary seems to concord with the Franciscan sources (represented in example first half m, pp. 332-335 2, p. 331, by the manuscript from Fribourg). An obvious example is the change of 477 Thirteenth Batholomaeus Liber epilogorum 134vb-135b Liber epilogorum, key. However, upon closer inspection one can detect differences in several passages, century, Tridentinus in gestis sanctorum De S. Cecilia, pp. especially in the choice of neumes. While the office for Cecilia from the Antoniana first half 350-354 does share great similarities with the Franciscan sources, it only perfectly coincides with the later source from Bergamo. The long melismas and verses in RR14 appear to be less susceptible to the influence of the German dialect than those in other sources The collection of manuscripts at the Antoniana is evidence of the extremely diverse from the same century and geographical area. In the other sources the versicles usually provenience of the friars who came to the basilica: in the thirteenth and fourteenth only consist of text, whereas in RR14 they are intoned on the chord F. centuries, they mainly came from France and England, and in the fifteenth century they came from Germany and the Low Countries. Their presence had a concrete 2.3 Cecilia as exemplum sanctitatis influence on the types of texts and iconographical materials acquired by the Library, in the hagiographical manuscripts on literary subjects which attests a graphical repertoire sometimes different for style from that Italian Cecilia is also mentioned as exemplum sanctitatis in the hagiographical manuscripts on (Giovè Marchioli 2003, pp. 671-690). At the same time, the library holdings themselves literary subject held by the Biblioteca Antoniana, including the Legenda aurea by Iacopo were not immune to the influence of the local liturgical tradition, as can be seen in de Varagine and two exemplars of the Liber Epilogorum by Bartolomeo da Trento. Such manuscript I-Psa107, where several chapters on the local saints of Padua Prosdocimo texts were used by preachers for preparing their homilies: the exemplars of the Legenda and Giustina have been added to the Legenda aurea. aurea were probably drafted for personal reading, although they attest some exempla of sanctity to enrich sermons to be read during the feasts of the liturgical year.8 The text was Conclusion composed just «pro utilitate […] pro informatione et eruditione omnium ecclesiasticarum The office for Saint Cecilia preserved in the Biblioteca Antoniana in Padua combines personarum» (Fleith 1991, p. 41). Table 2 presents a complete list of the hagiographical peculiar, local characteristics with a number of connections to a variety of sources. manuscripts on literary subjects held at the Antoniana containing episodes of the life While showing only sporadic concordances with other coeval sources from northern and passio of Cecilia (Iacopo da Varazze 2007; Bartolomeo da Trento 2001). and central Italy, the source from the Antoniana very often matches earlier sources from eleventh- and twelfth-century Aquileia and Ravenna. The most interesting 8. Mazzon 2003. On the interesting matter of the study of the books in the Franciscan Order, see Severino Polica (1976, pp. connections are the persuasive concordances with a group of about ten thirteenth- 373-413); on the libraries of the Mendicant Orders see Bartoli Langeli (1944, pp. 283-305); for Padua see Giovè Marchioli (2002, pp. 131-141). century Franciscan sources, antiphonaries and breviaries of central Italian provenience

332 333 CANTUS PLANUS 2014 MORANDI — THE OFFICE FOR SAINT CECILIA

(now scattered in various libraries worldwide) that also contain the office for Saint Appendix Anthony of Padua, composed by Julian of Speyer. By the time the office for Cecilia was composed, these Franciscan sources were already likely well-known at the basilica of List of abbreviations used in tables Padua, even if by virtue of the office for Saint Anthony alone. This would explain the 1ant = first antiphon 3 vrs = third verse r = recto peculiar choice of chants and melodies for the office for Cecilia, which lack confirmation 1nc = first Nocturn 3ant = third antiphon ra = recto higher 1rsp = first responsory 3rsp = third responsory rb = recto lower elsewhere. It is also possible that earlier sources from Aquileia and Ravenna served as 1vrs = first verse 4 ant = fourth antiphon rsb = short responsory the only models for the composition of the central Italian Franciscan sources. What is 1vs = first Vesper abd = ad benedictus antiphon rsp = responsory clear is that the office for Saint Cecilia had a strong influence on the composition of the 2ant = second antiphon act = action r-v = recto and verso 2nc = second Nocturn ain = invitatory v = verso office for Cecilia in the fifteenth-century source from Bergamo, as shown by frequent 2rsp = second responsory amg = ad magnificat antiphon va = verso higher similarities in their melodic and neumatic choices. This also demonstrates that the 2vrs = second verse ant = antiphon vb = verso lower office for Saint Cecilia preserved at the Biblioteca Antoniana is not merely a witness of 2vs = second Vesper laud = Lauds vrs = verse the confluence of various early and coeval liturgical traditions, but that it served as a 3 nc = third Nocturn nc = Nocturn vsc = versicle model for a later source, transmitting its peculiar lectio into the non-Franciscan realm of the Consortium of Charity of Santa Maria Maggiore in Bergamo.9

Bibliography

Abate Giuseppe (1950), Il libro e le biblioteche, Atti del I congresso bibliologico francescano, Roma: Pontificia Biblioteca Antoniana.

– (1935), ‘Le Constitutiones Generales Antique dei Frati Minori nella redazione assisiana del 1279’, Miscellanea Francescana 24, pp. 1-65.

Abate Giuseppe - Luisetto Giovanni (1975), Codici e manoscritti della Biblioteca Antoniana, 2 vols., Vicenza: Neri Pozza (Fonti e studi per la storia del Santo a Padova Fonti 1-2).

Bartoli Langeli Attilio (1944), I libri dei frati. La cultura scritta dell’Ordine dei Minori in Emanuela Prinzivalli ed., Francesco d’Assisi e il primo secolo di storia francescana, Torino: Einaudi, pp. 283-305 (Biblioteca Einaudi 1).

Bartolomeo da Trento (2001), Liber epilogorum in gestis sanctorum, ed. Emore Paoli, Firenze: Sismel Edizioni del Galluzzo (Edizione Nazionale dei Testi Mediolatini 2).

Canova Mariani Giordana (2010), ‘I manoscritti miniati della Biblioteca Antoniana. Nuove riflessioni sulla genesi della raccolta’, Il Santo 50/2-3, pp. 380-400.

Chiodi Luigi (1965), Guida agli archivi di antico regime di Bergamo e provincia nel 7° centenario di fondazione della Misericordia Maggiore di Bergamo, Bergamo: Centro studi Archivio Bergamasco.

Collamore Lila (2000), Aquitanian Collections of Office Chants: A Comparative Survey, Ph. D. dissertation, The Catholic University of America, Washington.

Ferreira Manuel P. (1997), Music at Cluny: the Tradition of Gregorian Chant for the Proper of the Mass. Melodic Variants and Microtonal Nuances, Ph. D. dissertation, Princeton University, Princeton.

Fleith Barbara (1991), Studien zur überlieferungsgeschichte der lateinischen Legenda aurea, Bruxelles: Société des Bollandistes.

Giovè Marchioli Nicoletta (2002), Circolazione libraria e cultura francescana nella Padova del Due e Trecento in Laura Gaffuri - Riccardo Quinto eds., Predicazione e Società nel Medioevo: riflessione etica, valori e modelli di comportamento. Atti del xii Medieval Sermon Studies Symposium (Padova 9. ia The Congregation of the Misericordia Maggiore (M ) was founded in Bergamo in 1265 by a group of laypeople led by 14-18 luglio 2000), Padova: Centro Studi Antoniani, pp. 131-141. the Dominican preacher, Pinamonte da Brembate, who established the ten chapters constituting its rules. The main aim of the Congregation is to provide charity and assistance for the poor, the sick, orphans and widows. The Mia’s Archive holds a vast archive of historical documents, including three printed books (cinquecentine), five manuscripts and fourteen choral – (2003), ‘Copisti, sottoscrizioni e scritture nei manoscritti della Biblioteca Antoniana’, Il Santo 43, books (Chiodi 1965, pp. 3-96). pp. 671-690.

334 335 CANTUS PLANUS 2014

Stefania Roncroffi Grier James (1995), ‘Roger de Chabannes (d. 1024), Cantor of Saint Martial de Limoges’, Early Music History 14, pp. 53-119. Antichi testimoni dell’ufficio di san Prospero e tradizione Hourlier Jacques (1959), ‘Le Bréviaire de Saint-Taurin: Un livre liturgique clunisien a l’usage de l’Échelle- Saint-Aurin (Paris, B. N. lat. 12601)’, Études grégoriennes 3, pp. 163-173. nei codici reggiani

Humphreys Kenneth William (1964), The Book Provisions of the Medieval Friars 1215-1400, Amsterdam: Erasmus Booksellers.

Iacopo da Varazze (2007), Legenda Aurea, ed. Giovanni Paolo Maggioni, Firenze: Sismel Edizioni del Galluzzo (Edizione Nazionale dei Testi Mediolatini 20) – Milano: Biblioteca Ambrosiana. Le più antiche attestazioni di un ufficio con notazione musicale dedicato a san Prospero, patrono di Reggio Emilia, sono tramandate da due preziosi codici, entrambi Lamothe Donat R. - Constantine Cyprian G. eds. (1986), Matins at Cluny for the Feast of Saint Peter’s Chains, London: Plainsong and Mediaeval Music Society. della seconda metà dell’undicesimo secolo: I-VEcap XCVIII (92) e V-CVbav lat. 7018. Il codice veronese è un antifonario di grande interesse, tanto che è stato assunto da René- eroquais Les Bréviaires: manuscrits des Bibliothèques publiques de France L Victor (1934), , 5 vols., Paris- Jean Hesbert (1963-1970, 1) nella sua monumentale opera Corpus Antiphonalium Macon: Protat Frères. Offici, insieme ad altri cinque manoscritti, come modello di riferimento per la liturgia Luisetto Giovanni (1999), ‘La Biblioteca Antoniana del Convento del Santo di Padova’, Notiziario del cursus romano.1 Costituito da 267 carte, presenta una notazione adiastematica Bibliografico 32, pp. 2-66. con neumi in campo aperto diffusa nell’Italia settentrionale, e alcune parti in nota- Maranesi Pietro (2000), Nescientes Litteras. L’ammonizione della Regola Francescana e la questione degli zione nonantolana. Nel santorale si rinvengono tre uffici propri: per san Zeno, per studi nell’Ordine (secoli XIII-XVI), Roma: Edizioni dell’Istituto Storico dei Cappuccini (Biblioteca i santi Fermo e Rustico e per san Prospero. La presenza degli uffici dei santi protet- seraphico-capuccina 61). tori (patrono e co-patroni) di Verona inducono a propendere per questa città come Mazzon Antonella (2003), Manoscritti agiografici latini conservati a Padova, Firenze: Sismel Edizioni del luogo di produzione e utilizzo del manoscritto. Tuttavia la festa di san Zeno è stata Galluzzo (Quaderni di Hagiografica 2). inserita posteriormente nel codice,2 mentre la solennità di san Prospero risulta del

Morandi Nausica (2006), ‘L’Ufficio di Santa Cecilia nelle fonti italiane. Saggio di edizione e analisi tutto inconsueta per il santorale veronese. Di fatto sia Prospero, sia Zeno, sia Fermo comparativa’, Rivista Internazionale di Musica Sacra 27/1, pp. 71-151. e Rustico hanno goduto di un culto abbastanza diffuso e, anche se quella di Verona

Nardini Roberto (2008), ‘La Ratio formationis dei benedettini di Monte Oliveto’, L’ulivo 38, pp. 182-197. rimane l’ipotesi più accreditata, non si esclude che il manoscritto sia stato prodotto per un centro dell’Italia settentrionale non troppo lontano da Verona, dove peraltro Re Caterina (1929), ‘I Chiostri del Convento del Santo’, Il Santo 2, pp. 170-217. in seguito fu probabilmente utilizzato.3 V-CVbav lat. 7018 è un breviario messale

Schaff Philip (1997), History of the Christian Church, Oak Harbor: Logos Research Systems. con notazione musicale più o meno coevo al codice veronese. Consta di 228 carte, ma si compone in realtà di due manoscritti incompleti riuniti insieme. Presenta un Severino Polica Gabriella (1978), Libro, lettura, lezione negli Studia degli Ordini mendicanti (sec. XIII) in contenuto molto vario in cui si colloca il proprio dei santi dalla festa di Tiburzio e Centro studi sulla spiritualità medioevale ed., Le scuole degli Ordini mendicanti. Atti del XVII Convegno del Centro di Studi sulla spiritualità medievale (Todi, 11-14 ottobre 1976), Todi: Accademia Valeriano fino all’ottava di sant’Andrea. Vi si trovano culti peculiari per alcuni santi Tudertina, pp. 373-413. venerati in ambito francese, e propri dell’area italiana.4 Tra questi, oltre a Procolo e a Donato vescovo di Arezzo, spicca il nome di Prospero, cui è dedicato un ufficio ampio Steiner Ruth (1933), Marian Antiphons at Cluny and Lewes in Susan Rankin - David Hiley eds., Music in the Medieval English Liturgy: Plainsong and Mediaeval Music Society Centennial Essays, Oxford: e completo. Il santo è definito «pastor noster», e tale appellativo, unitamente al rilievo Clarendon Press, pp. 175-204. conferito all’ufficio, ha indotto Pierre Salmon (1964, 7, p. 331; 1968, p. 183) a ritenere

1. Per l’identificazione di antifone e responsori il repertorio sarà citato come Cao seguito dai numeri da 1001 a 5517 per il vol. 3: Invitatoria et antiphonae (1968), e da 6001 a 8455 per il vol. 4: Responsoria, versus, hymni et varia (1970).

2. La festa di san Zeno, inserita posteriormente alla confezione del codice, è posizionata tra l’Invenzione della croce (3 maggio) e l’Ascensione. Si può riferire dunque alla festa della Traslazione del santo celebrata il 21 maggio (Hesbert 1963, 1, pp. xxii-xxiii), che tuttavia è piuttosto rara nei manoscritti anteriori al quattordicesimo secolo, oppure alla festa della Deposizione di san Zeno che invece è tipica del santorale veronese e cade il 12 aprile (Petterlini 1997, pp. 304-305). Anche se le carte aggiunte, nel rispetto del calendario, avrebbero dovuto essere posizionate prima dell’Invenzione della croce, l’inserimento è collocato in un punto del codice in cui la fine di un ufficio corrisponde alla fine di un fascicolo (Petterlini 1993-1994, pp. 18-19).

3. Per una descrizione del manoscritto cf. Polloni (2012, p. 521) con bibliografia precedente.

4. Per una descrizione del manoscritto cf. Bannister (1913, pp. 90-91, scheda n. 261 riferita all’illustrazione no. 51).

336 337 CANTUS PLANUS 2014 RONCROFFI — TESTIMONI DELL’UFFICIO DI SAN PROSPERO NEI CODICI REGGIANI

il manoscritto di provenienza reggiana. Successivamente Cesarino Ruini (1990, pp. inseriti in luogo dei responsori Iste est vere Israelita V. Dedit ei (Cao 7005) e O beati viri 130-131; 2014a, pp. 103-104) ha dimostrato che il codice fu confezionato per Pistoia: Prosperi V. Cuius vivere Christus (Cao 7255). la localizzazione dell’area di provenienza è attestata dalla data della festa indicata sul manoscritto, il 25 novembre, come è tradizione in Toscana e non il 24 come nel calen- L’ufficio descrive con toni celebrativi la biografia e la figura del santo, a cui attribuisce dario reggiano, ma soprattutto dalla lista dei versetti alleluiatici delle domeniche dopo un’origine spagnola, secondo una tradizione diffusa nel Medioevo. Abbandonata la Pentecoste, corrispondente a quella di I-PS C 119 e C 120. I due manoscritti, databili ai patria, Prospero è guidato dal volere divino a Reggio, dove fu scelto come vescovo: «in secoli undicesimo/dodicesimo, contengono inoltre la festa per san Prospero, al quale urbe quam dicunt emiliam beatus prosper tenuit cathedram».9 È definito «facie ange- era dedicata, almeno dal 1137, una chiesa parrocchiale in città (Bergamaschi 2014, licus, sermone fecundus, habitu humilis, animo prudentissimus». Fu attivo contro i pp. 177-180). Il culto per san Prospero infatti ebbe un’ampia diffusione soprattutto in movimenti ereticali e dopo la morte compì numerosi miracoli, sanando gli ammalati Emilia (da Piacenza a Imola), ma anche nel sud della Lombardia, in Liguria, nella che accorrevano al suo sepolcro. Il popolo reggiano lo invoca a protezione della città, Toscana centro-settentrionale fino a Siena e in Umbria (Costi 2012, p. 106). Di fatto si come mediatore presso Dio per ottenere grazie e la remissione dei peccati. A Reggio tratta dei territori matildici, di cui Reggio si trova al centro, più o meno equidistante in in suo onore si celebravano il dies natalis il 25 giugno, e la Traslazione delle spoglie il linea d’aria da Pistoia e Verona, quest’ultima peraltro esclusa di poco dal dominio dei 24 novembre. In I-VEcap XCVIII (92) l’ufficio è inserito ai fogli 209-211, tra le feste dei Canossa. E all’epoca matildica risalgono le due antiche fonti che tramandano l’ufficio santi Clemente (23 novembre) e Andrea (30 novembre). Si riferisce dunque alla data di san Prospero in una versione simile fra loro anche se non identica. In entrambi i di novembre, anche se la rubrica recita In Natale Sancti Prosperi. In V-CVbav lat. 7018 manoscritti l’ufficio comprende un’antifona al Magnificat per i vespri, un invitatorio l’ufficio, contenuto ai fogli 154v-156v, è preceduto da una rubrica che indica la data per il mattutino, quindi tre antifone e tre responsori per ciascuno dei tre notturni, nel settimo giorno precedente le calende di dicembre. Si riferisce dunque alla festa cinque antifone per le lodi e un’antifona al Benedictus. In I-VEcap XCVIII (92) al ter- della Traslazione, celebrata il 25 novembre secondo il calendario diffuso in Toscana. mine dell’ufficio si aggiunge Hodie beatissimus Prosper, antifona al Magnificat, che in In VEcap XCVIII (92) le rubriche sono limitate a in vigilia, in die e in laudibus, e sono V-CVbav lat. 7018 è collocata ai primi vespri nella stessa posizione di Salva plebem, assenti gli incipit dei salmi previsti per le lodi. In V-CVbav lat. 7018 si rinvengono le assente in V-CVbav lat. 7018. Inoltre in quest’ultimo codice è inserita anche l’anti- rubriche per i vespri, i notturni e le lodi, ma mancano le melodie per tutti i versetti fona Laudemus Deum (Cao 1098) tratta dalla festa di San Martino, da cantarsi «super dei responsori. Essendo un breviario-messale contiene anche le letture per notturni Venite», che non si trova invece in I-VEcap XCVIII (92). Prendendo a modello il testo e, al termine dell’ufficio, gli incipit dei canti della messa con un’orazione conclusiva. di quest’ultimo codice, edito interamente da Hesbert (1963, 1, pp. 339 e 341), si nota Entrambi i codici sono scritti in una notazione adiastematica che unisce elementi pro- che la maggior parte delle antifone e dei responsori sono comuni ai due manoscritti pri di notazioni italiane con influenze oltremontane e si inserisce nel contesto della ma, ad eccezione delle lodi, ne viene stravolto l’ordine di esecuzione. Si riscontrano grande varietà di tradizioni grafiche presenti nel nord della penisola nell’undicesimo inoltre varie modifiche nell’abbinamento del versetto al responsorio,5 l’inserimento secolo (Baroffio 2008, p. 166). Le due notazioni, pressoché coeve, si presentano abba- di due versetti nuovi,6 ma soprattutto in V-CVbav lat. 7018 sono presenti due testi che stanza differenti fra loro. In V-CVbav lat. 7018 si osservano neumi stretti, dagli angoli non compaiono in I-VEcap XCVIII (92): il secondo responsorio del secondo notturno arrotondati e una leggera inclinazione verso destra di quasi tutte le figure. In I-VEcap Sancte Prosper V. Dedit ei dominus7 risultato inedito, e il terzo responsorio del terzo XCVIII (92) i neumi mostrano un tratto ben marcato, spesso sono angolati, verticali e notturno O quam beatus V. Securus habitationem, il cui utilizzo è attestato in alcuni abbastanza spaziati. Vi si nota la presenza di due forme diverse per rappresentare il manoscritti prodotti tra l’undicesimo e il quattordicesimo secolo.8 I due canti sono pes, la clivis e lo scandicus. Il pes è quadratus con un unico tratto in direzione ascen- dente, ma, seppur raramente, si trova anche il pes disgregato, costituito da un punctum

5. In I-VEcap XCVIII (92) il versetto Cuius vivere Christum è abbinato a Felix tanti viri (Cao 6726), in V-CVbav lat.7018 a Vir e da una virga. La clivis è nella forma di un pes rovesciato quando è preceduta da una igitur (Cao 7895). Il versetto Cuius venerabile corpus presenta alcune diversità di redazione nei due manoscritti. nota più alta o all’unisono, è arrotondata quando è preceduta da una nota più bassa. 6. In V-CVbav lat. 7018 è inedito il versetto Ecce iste est vir prudens beatus Prosper, abbinato al responsorio Cumque ab Nettamente prevalente la prima tipologia, con un’incidenza di circa sei volte superiore Hispania, erroneamente trascritto dal copista Cumque habiit Ispania. Il versetto del responsorio Felix tanti viri (Cao 6726) è Ipse quoque tamquam grex (Cao 6796), tratto dall’ufficio di san Maurizio e in parte rielaborato. rispetto all’altra. Lo scandicus è rappresentato da una prima tipologia più diffusa costi- 7. Sancte Prosper pastor noster tuo gregi subveni ut qui nostra fragilitate destruimus tuis precibus deo reo reformemur. tuita da un segno verticale somigliante a una virga, sotto il quale sono posti due punti, V. Dedit ei d[omi]n[u]s s[an]ctitatis stolam et constituit eum principem familiae suae. Ut qui (V-CVbav lat. 7018, fol. 155v). ma si ritrova anche lo scandicus con virga terminale, costituito da una virga preceduta 8. O quam beatus es pie confessor dei hodie namq[ue]chorus angelorum revertentem ad patriam suscepit intercede pro nobis. V. Securus habitationem istam reliquisti et usq[ue]ad solium visionis aeternae penetrasti. Inter[cede] (V-CVbav lat. da un pes. Il climacus è composto da un tratto orizzontale leggermente inclinato verso 7018, fol. 156). Il responsorio non compare nei più comuni repertori a stampa. È comunque documentato in undici fonti schedate nel database Cantus. Si trova impiegato nella liturgia del comune dei confessori, dei martiri, degli abati e delle feste di san Cuthberto, Adolfo, Marziale, Osvaldo e Orienzio (). quindicesimo secolo.

338 339 CANTUS PLANUS 2014 RONCROFFI — TESTIMONI DELL’UFFICIO DI SAN PROSPERO NEI CODICI REGGIANI l’alto, alla base del quale sono posizionati due punti, il secondo di questi ha spesso secoli e si riscontra in manoscritti realizzati per istituzioni religiose reggiane dal forma ripiegata verso il basso somigliante a una virgola. Si nota poi la presenza sia di dodicesimo al quindicesimo secolo. Il modello è quello del codice veronese: lo con- punctum che di tractulus, di cephalicus e ancus per indicare le liquescenze, dell’apo- ferma l’identica successione dei canti e l’assenza dei testi propri di V-CVbav lat. 7018. stropha e di vari neumi composti, nati dalla fusione di neumi semplici. Osservando La prima attestazione in ordine cronologico si trova in I-REm Reggiani C 408 (fine nel complesso questa notazione, propria dell’area veronese (Baroffio 2011, p. 211), del dodicesimo secolo - inizi del tredicesimo), in cui al foglio 75r-v, in una sezione emergono alcuni tratti in comune con la notazione di Metz, riportata nel graduale destinata a canti commemorativi, sono riportati il responsorio Cumque ab Hispania F-La 239 del decimo secolo, un’ascendenza verificata da David Hiley (1993, p. 351) (Cao 6379) e l’antifona Salva plebem (Cao 4688), in ‘nota romana’ su rigo a secco anche per i manoscritti di canto liturgico prodotti nell’area fra Novalesa, Asti, Brescia con linea rossa per il Fa e gialla per il Do (Ruini 2013, pp. 546-547). Un’altra fonte, e Mantova. In particolare le somiglianze più evidenti si rilevano nella clivis a forma di recentemente riscoperta, è fornita dal manoscritto GB-Mr L 74, un pregevole anti- pes rovesciato, nelle due tipologie di pes, e nel porrectus. Inoltre un altro segno tipico fonario datato dagli storici della miniatura agli anni settanta del secolo xiii, portato dei neumi di F-LA 239 è costituito da un punctum seguito da una clivis corrispondente in Inghilterra dall’antiquario Guglielmo Libri (Zanichelli 2008, p. 575-576). Lo stile al trigon, peraltro estraneo alla notazione metense. Anche il torculus sembra derivare delle miniature di scuola bolognese e la presenza di un lungo ufficio dedicato a san da questa famiglia neumatica, ma viene sottolineato il tractulus o punctum iniziale, Prospero fanno propendere per un’origine reggiana del manoscritto che presenta la cui se ne aggiunge uno finale. stessa successione di antifone e responsori di I-VEcap XCVIII (92). Il codice cronolo- gicamente successivo contenente l’ufficio di san Prospero è il I-REm Reggiani C 407, V-CVbav lat. 7018 presenta una tipologia di notazione dell’Italia settentrionale che un libro ordinario prodotto per la cattedrale di Reggio Emilia nel secolo quattordice- invece sembra derivare dai neumi propri dei manoscritti di San Gallo, con alcune simo, che riporta quindi solo gli incipit dei testi. Si tratta tuttavia di una fonte molto analogie con il codice bolognese I-Ra 123. Le somiglianze più evidenti si riscontrano importante, in quanto modello liturgico per i volumi realizzati successivamente, tre nella figura della clivis e del porrectus, entrambe con il primo elemento arrotondato. antifonari del quindicesimo secolo che tramandano l’intero ufficio in notazione neu- Anche il climacus, che si configura come una virga con tratto più spesso e ricurva matica quadrata: I-REm 17 A 136 (pp. 87-117), I-REsp C 2 (pp. 38-67) e I-REsp N (pp. verso il basso, seguita da due punti, mostra somiglianze con la notazione bolognese, 334-363). In quest’ultimo codice l’ufficio è previsto per la festa della Traslazione, così come lo scandicus, il quilisma-pes e le due tipologie di torculus. In alcuni casi mentre in I-REm Reggiani C 407 una nota a margine di seconda mano per questa l’ultimo punctum di una serie discendente è in forma di virgola e questo aspetto si ricorrenza prescrive «Officium de communi vel proprium». rileva anche in I-VEcap XCVIII (92). Le varie fonti concordano in linea generale con il manoscritto veronese sia per i testi Tabella 1. I neumi principali che per la melodia. In I-REsp N tuttavia sono stati introdotti tre nuovi brani poetici: il responsorio Gaude mater ecclesia (p. 350) e le antifone Gaude felix parens Hyspania (p. Codice punctum virga pes clivis scandicus climacus torculus porrectus 361) e O decus Hyspanie (p. 362), che sostituiscono rispettivamente il primo responso- I-VEcap XCVIII (92) rio del primo notturno, l’antifona al Benedictus e l’antifona al Magnificat dei secondi

vespri.10 La melodia si tramanda abbastanza stabilmente nelle fonti più recenti seppur V-CVbav lat. 7018 con lievi varianti, fenomeno del tutto consueto, soprattutto se si considera la distanza cronologica dal testimone più antico e la mancanza di una versione locale dell’ufficio. Le esigue modifiche alla linea melodica non seguono una tendenza comune poiché in alcuni casi si osservano note aggiunte nei manoscritti più recenti, in altri invece gli stessi risultano più scarni rispetto al codice veronese.11 I testi di questo antico ufficio I punti di maggior convergenza tra i due manoscritti si notano nel pes, nella clivis sono stati editi negli Officia sanctorum Patronorum et Protectorum urbis Regiensis, la arrotondata e nello scandicus. Tuttavia, alla concordanza dei testi (seppur con le cui prima edizione venne stampata a Bologna nel 1515 (riprodotta in Ruini 2014a). Si varianti segnalate) e delle melodie, che paiono seguire nella loro vaga insita dia- tratta di un libretto di piccolo formato, destinato alla devozione privata, in cui sono stemazia un generale andamento comune, si contrappone l’aspetto della notazione, che in I-VEcap XCVIII (92) sembra condizionata da influssi provenienti dal nord-est della Francia, in V-CVbav lat. 7018 appare legata a I-Ra 123, a conferma delle rela- 10. I testi del responsorio e delle due antifone sono stati pubblicati da Ruini (1990 p. 139), le musiche da Roncroffi (2000, zioni dell’ambiente bolognese con l’area umbro-toscana (Ruini 2011, pp. 241-242). pp. 110-111). La tradizione dei testi e delle melodie di questi due antichi testimoni si mantiene nei 11. La tradizione dell’ufficio nei codici conservati a Reggio Emilia è stata descritta in Roncroffi (2014, pp. 109-119).

340 341 CANTUS PLANUS 2014 RONCROFFI — TESTIMONI DELL’UFFICIO DI SAN PROSPERO NEI CODICI REGGIANI raccolti i testi senza le musiche per la celebrazione delle ore canoniche.12 Ad eccezione Bibliography del responsorio Laudemus dominus V. Ante cuius (Cao 7082) si rinvengono qui tutti i canti presenti in I-VEcap XCVIII (92), riportati integralmente e nella stessa succes- Bannister Henry Marriott (1913), Monumenti Vaticani di paleografia musicale latina raccolti ed illustrati, sione (Ruini 2014b, pp. 324-331), segno di una tradizione consolidata e di un culto Lipsia: Harrassowitz. sentito che, almeno in ambito reggiano, si è ben radicato nel tempo. Baroffio Giacomo (2008), in Calzona 2014, pp. 164-175.

– (2011), Music Writing Styles in Medieval Italy in John Haines ed., The Calligraphy of Medieval Music, Turnhout: Brepols (Musicalia Medii Aevi I), pp. 101-124.

Bergamaschi Gianni (2014), Culti transappenninici in Toscana: testimonianze liturgiche e agio-topomastiche in Renato Stopani - Fabrizio Vanni, Tra due romee: storia, itinerari e cultura dei pellegrinaggi in Val d’Orcia. Atti del Convegno (Monticchiello, Pienza, e Abbazia di Spineto, Sarteano, 7-8 giugno 2013), Firenze: Centro Studi Romei, pp. 157-204.

Calzona Arturo ed. (2014), Matilde e il tesoro dei Canossa tra castelli, monasteri e città: catalogo della mostra, Cinisello Balsamo: Silvana.

Costi Giovanni (2012), San Prospero vescovo e patrono di Reggio (461-483) in Giovanni Costi - Giuseppe Giovanelli eds., Storia della Diocesi di Reggio Emilia e Guastalla, 3 vols., Brescia: Morcelliana, 1: Dalle origini al Medioevo, pp. 87-131.

Hesbert René-Jean (1963-1970), Corpus Antiphonalium Officii, 6 vols., Roma: Herder (Rerum Ecclesiaticarum Documenta Cura Pontifici Athenaei Sancti Anselmi de Urbe Edita. Series maior: Fontes 7-12).

Hiley David (1993), Western Plainchant: a handbook, Oxford: Clarendon.

Petterlini Elena (1993-1994), ‘Per lo studio del santorale veronese: fonti liturgico-musicali della Biblioteca Capitolare (secoli IX-XIV)’, Rassegna veneta di studi musicali 9-10, pp. 5-57.

– (1997), Il santorale marciano e il Santorale della chiesa di Verona in Giulio Cattin ed., Da Bisanzio a San Marco. Musica liturgia, Venezia: Levi - Bologna: Il Mulino (Quaderni di «Musica e Storia» 2), pp. 301-313.

Polloni Susanna (2012), I più antichi codici liturgici della Biblioteca Capitolare di Verona (secoli V-XI). Catalogo descrittivo, Verona: Archivio Storico Curia Diocesana (Studi e Documenti di Storia e Liturgia 45) pp. 513-522.

Roncroffi Stefania (2000), Il fondo musicale della Basilica di San Prospero a Reggio nell’Emilia, Firenze: Olschki (Historiae Musicae Cultores 87).

– (2014), Le fonti musicali dell’ufficio di san Prosperoin Ruini 2014, pp. 109-119.

Ruini Cesarino (1990), Caratteri peculiari del canto liturgico a Reggio Emilia? in Angelo Pompilio - Donatella Restani - Lorenzo Bianconi - F. Alberto Gallo eds., Atti del XIV congresso della Società Internazionale di Musicologia: Trasmissione e ricezione delle forme di cultura musicale, 2 vols., Torino: Edt; 2: Study Sessions, pp. 130-149.

– (2011), Il codice Angelica 123: Musica e politica tra i secoli XI e XII in Giovanni Feo ed., Bologna e il secolo XI. Storia, cultura, economia, istituzioni, diritto, Bologna: Bononia University Press, pp. 239- 252.

– (2013), «Nota Romana in Aemilia». Documenti sulla diffusione della notazione dell’Italia centrale nella diocesi di Reggio Emilia in Barbara Haggh-Huglo e Debra Lacoste eds., Papers Read at the 15th Meeting of the IMS Study Group Cantus Planus (Dobogókő/Hungary, 2009, August 23-29), Lions Bay, Bc: The Institute of Mediaeval Music, pp. 543-556. 12. Il libretto venne ristampato nel 1567, nel 1575 e nel 1600 e rimase immutato fino al 1626, quando dalla Tipografia della Reverenda Camera Apostolica di Roma usciva una edizione degli uffici che prescriveva il ricorso ai formulari del comune dei santi (Ruini 2014a, p. 123). – ed. (2014), Vere Dignum. Liturgia, musica, apparati, Bologna: Pàtron (Ecclesia regiensis, 4).

342 343 CANTUS PLANUS 2014

Diego Toigo – (2014a), Antichi testi liturgici reggiani: musica e canto in Ruini 2014, pp. 103-130. – (2014b), Gli uffici propri dei santi reggianiin Ruini 2014, pp. 321-366. Il canto del Passio presso alcuni ordini religiosi

Salmon Pierre (1964), Un bréviaire-missel du XIe siècle. Le manuscrit vatican latin 7018 in Mélanges Eugène in fonti italiane tardive Tisserant, 7 vols., Città del Vaticano: Biblioteca Apostolica Vaticana (Studi e Testi 231-237), vol. 7, pp. 327-343.

– (1968), Les manuscrits liturgiques latins de la Bibliotheque Vaticane, 3 vols., Città del Vaticano: Biblioteca Apostolica Vaticana, 1: Psautiers, Antiphonaires, Hymnaires, Collectaires, Breviaires (Studi e Testi 251), pp. 182-183. Essendo ormai chiaro il quadro delle intonazioni monodiche della Passione finora emerse in area italiana in edizioni e manoscritti di età medievale e rinascimentale,1 Zanichelli Giuseppa (2008), Antiphonarium: scheda n. 136 in Arturo Calzona ed., Matilde e il tesoro dei Canossa tra castelli, monasteri e città: catalogo della mostra, Cinisello Balsamo: Silvana, pp. 575-576. risulta ora più agevole allargare l’indagine ai secoli successivi; anzi, l’esame delle testimonianze più recenti può contribuire a precisare i contorni generali del disegno e a meglio valutare il grado di popolarità di certe intonazioni più antiche. Per un’adeguata contestualizzazione di queste nuove testimonianze, devo premettere che lo studio delle intonazioni di epoca medievale e rinascimentale ha consentito di stabilire che in area italiana esisteva, fra i secoli tredicesimo e sedicesimo, una tradizione prevalente di origine romano-francescana, attestata non solo in cantorini manoscritti tardo-medievali, ma anche in edizioni cinquecentesche di cantorini e di rituali;2 il suo tratto distintivo è la spiccata tendenza a utilizzare soluzioni espressive, di sapore talvolta drammatico. Di questa intonazione propongo subito in trascrizione tre luoghi caratteristici, ricavati dal ms. I-Bu 2893 della Biblioteca Universitaria di Bologna:3 si veda innanzitutto l’initium del Cristo (esempio 1), che sale di una sesta, dal Fa al Re, per mezzo di un salto di terza; quindi la melodia per le ultime parole di Cristo sulla croce nella versione secondo Giovanni (esempio 2); infine, il tratto omologo dalla Passione secondo Matteo, la cosiddetta melodia-Eli (esempio 3), contraddistinta dall’inconfondibile doppio vocalizzo che amplifica e drammatizza le due invocazioni: «Eli, Eli», «Deus meus, Deus meus».

Esempio 1. Bologna, Biblioteca universitaria, manoscritto 2893, p. 263

1. In Toigo (2017) si trova un disegno complessivo delle tradizioni monodiche diffuse in area italiana fino al sedicesimo secolo, ricavato dalle fonti fino ad oggi emerse; segnalo inoltre: Toigo (2011a; 2011b; 2012a; 2012b; 2012c; 2015). Rimangono ancora utili Baroffio e Antonelli (1986) e Stäblein (1962).

2. L’elenco di tutte queste fonti si trova in Toigo (2017). Qui mi limito a menzionarne i seguenti cantorini manoscritti: I-Bu ms. 2893, quattordicesimo e quindicesimo secolo; I-TRfeininger ms. FC 133, quindicesimo secolo; I-TRsf ms. 312, quindicesimo secolo. Quanto alle fonti a stampa, oltre al noto cantorino romano del 1513 (Compendium musices 1513) e alle sue cinque ristampe, si vedano anche le quattordici edizioni del Familiaris clericorum liber, che uscì per la prima volta nel 1517 col titolo di Familiaris clericorum.

3. Su questo manoscritto cf. Stäblein (1962, coll. 891-894 e 895).

344 345 CANTUS PLANUS 2014 TOIGO — IL CANTO DEL PASSIO IN FONTI ITALIANE TARDIVE

Esempio 2. Bologna, Biblioteca universitaria, manoscritto 2893, p. 268 sede ho potuto sottolineare come le formule di Passione della tradizione monastica cassinese contenessero, in pieno Cinquecento, la melodia-Eli romano-francescana con doppio vocalizzo;8 ora vorrei considerare altre fonti di vari ordini religiosi, allo scopo di verificare se e quanto l’intonazione romano-francescana tardo-medievale vi abbia lasciato traccia.

Il Directorium carmelitano Una testimonianza seicentesca della Passione si legge in un Directorium chori (1614) Esempio 3. Bologna, Biblioteca universitaria, manoscritto 2893, p. 266 di origine carmelitana, che uscì a Napoli per iniziativa del priore generale padre Sebastiano Fantoni.9 Nell’epistola iniziale, rivolta «ai diletti figli Carmelitani», il priore informa di aver voluto uniformare il modus cantandi dell’Ordine, che non era celebrato secondo una stessa regola nei vari conventi; per tale operazione di rinnovamento egli dice di aver goduto dell’aiuto di Arcangelo Paoli, allora maestro dei novizi ad Aversa, il quale «non modo ad hunc laborem suscipiendum visus fuit idoneus, sed (quod gratius et optabilius contigit) paucis ab hinc annis Directorium Identica all’intonazione del manoscritto bolognese è quella contenuta in un altro Chori non mediocri labore ac studio confecerat».10 Alle pp. 65-67 il volumetto presenta cantorino coevo, D-Bds Mus. 40562, di cui oggi rimane solo la riproduzione in le formule di Passione secondo il Vangelo di Matteo. Vi si riconosce subito il melisma microfilm.4 Numerose attestazioni della sola melodia-Eli con doppio vocalizzo si romano-francescano che apre gli interventi del Cristo (lo stesso dell’esempio 1). Un trovano in messali ed evangeliari manoscritti quattrocenteschi, trecenteschi, e altro elemento importante è costituito dalla corda di recita destinata alle turbe, qui anche di fine Duecento, se consideriamo il manoscritto della Biblioteca del Comune collocata sul Mi: non si tratta di una semplice preparazione al Fa, come di solito e dell’Accademia Etrusca di Cortona (I-CT 16).5 Fra tutti questi testimoni, in accade nella formula di interrogatio, giacché il Mi permane anche quando la melodia particolare, spicca il manoscritto V-CVbav Arch. S. Pietro B 69 (Biblioteca Apostolica trova riposo nel punctum. L’intonazione, dunque, si discosta dalla tradizione più Vaticana), un messale che fu in uso presso la cappella papale in pieno Quattrocento:6 tipicamente italiana, che riserva alle turbe la corda di Fa, ma non costituisce un il fatto che questo codice contenga la melodia-Eli nella formulazione romano- caso isolato, considerando che agli inizi del Cinquecento anche l’intonazione francescana, permette di ipotizzare che a Roma, prima del sedicesimo secolo, il monastica cassinese era modulata sui toni di Sol, Do e Mi, come testimonia, ad canto del Passio fosse intonato secondo la tradizione romano-francescana, e non esempio, il codice 783.2 L712 della University Library dell’Università dell’Illinois.11 secondo il noto modello del Cantus ecclesiasticus Passionis mandato alle stampe Interessante, nel Directorium, è anche la melodia per le ultime parole di Cristo sulla da Giovan Domenico Guidetti nel 1586. Ribadisco, anzi, che l’intonazione romano- croce, intonata in ambito acuto, espressiva e patetica nel suo andamento cantabile francescana dei cantorini differisce decisamente da quella proposta dal Guidetti (figura ).1 Una formula molto simile a questa, in realtà, era già apparsa almeno un e ristampata più volte nel corso del Seicento.7 Quando e perché a un certo punto secolo prima nelle due edizioni del Familiaris clericorum liber pubblicate a Venezia a Roma sia emersa quell’intonazione non è ancora chiaro. Ora, nonostante l’opera da Gregorio de’ Gregori nel 1517 e nel 1525;12 dunque, non si tratterebbe di una del Guidetti si andasse diffondendo grazie alle ristampe seicentesche, pare che la realizzazione di ambito carmelitano, quanto piuttosto di una formula di tradizione melodia romano-francescana abbia continuato ad esercitare un certo influsso sulle romano-francescana.13 Il Directorium chori carmelitano fu più volte ristampato fra intonazioni di epoca più tarda, e non solo in ambito secolare (Toigo 2015). Già in altra 8. Per tutto questo rimando a Toigo (2017, capitolo 2, paragrafo 1.3 e capitolo 4, paragrafo 2). 9. Circa l’origine e la storia dell’Ordine cf. Catena (1975).

4. Il codice, un tempo conservato appunto a Berlino, è andato perduto nel 1945 ed è oggi consultabile in microfilm presso 10. La lettera è datata al 20 novembre del 1613, e figura subito dopo il frontespizio, prima dell’indice. il Musikwissenschaftliches Seminar dell’Università di Erlangen (Stäblein 1962, coll. 891-894). 11. US-U ms. 783.2 L712, inizio del secolo sedicesimo (d’ora in poi ms. ‘Illinois’). Per un quadro complessivo delle fonti 5. Una breve descrizione di questo messale è presente in Sorbelli (1991, p. 10). Per altre notizie e per un elenco di tutte le cassinesi della Passione, rimando a Toigo (2017, capitolo 2, par. 1). fonti sopra citate, rimando a Toigo (2017, capitolo 6). 12. Cf. nota 2. Questo rituale, prodotto in ambito francescano, ebbe poi grande fortuna e fu ristampato altre dodici volte nel 6. Sulla provenienza cf. Roth (1983, p. 264, nota 46). corso del Cinquecento (Toigo 2017, capitolo 1, paragrafo 4).

7. Ristampe: 1615, 1627 (Roma: Andrea Fei), 1644 (Roma: Antonio Poggioli), 1689 (Roma: Campana). La stessa intonazione 13. Una variante della stessa melodia si riconosce anche in V-CVbav ms. Ott. lat. 221, fol. 61v del Quattrocento. La formula sarà ripresa in pieno Ottocento da Pietro Alfieri e dall’editore Pustet di Ratisbona. Per una trattazione delle complesse fu annotata nel corso del secolo seguente sul margine inferiore del fol. 61v (Baroffio e Antonelli 1986, p. 31, lettera i, con vicende che hanno interessato l’intonazione romana nel corso dei secoli, rimando a Toigo (2017, capitolo 6). breve trascrizione a p. 32, esempio 1, lettera i).

346 347 CANTUS PLANUS 2014 TOIGO — IL CANTO DEL PASSIO IN FONTI ITALIANE TARDIVE

Seicento e Settecento su iniziativa di altri priori generali, senza che le formule di Un testimone francescano Passione siano state modificate.14 A Venezia, presso l’Archivio della Provincia Veneta dei Frati minori, è conservato un cantorino francescano compilato nel 1715 da Giovanni da Bassano; nelle note destinate «ad cantores et lectores», il frate attesta di aver ricavato il contenuto «ex correctis exemplaribus, et praecipue ex antiquo Cantorino, impresso Venetiis per dominum Antonium de Giunta Florentinum»15 (con evidente allusione alla nota edizione giuntina del 1513).16 Le formule (fols. 29-31v) sono ricavate dalla Passione secondo Giovanni e trasportate di una quarta verso il basso; tolta la trasposizione, i toni di recitazione sono quelli tipici della tradizione italiana (Sol/Do/Fa); l’initium del Cristo e le ultime parole di Cristo sulla croce (figura 2) rispecchiano i modelli romano-francescani degli esempi 1 e 2. In particolare, il melisma sul «Consumatum est» reinterpreta il modello originario in modo libero, offrendo un esempio interessante di quella commistione semiografica fra gruppi ritmicamente connotati e gruppi neumatici tipici del canto piano.

Figura 1 Directorium chori, Napoli, Carlino, 1614, p. 67 Processionali domenicani Per quanto riguarda la tradizione domenicana, non pare che le formule di Passione contenute nel ‘prototipo’ dell’Ordine, il noto Gros livre conservato a Roma presso la Basilica di Santa Sabina,17 abbiano influenzato la tradizione successiva, in Italia. Non lo si può tuttavia asserire con sicurezza, essendo troppo ampio il divario temporale che separa il prototipo medievale dalla prima attestazione a me nota di una Passione con notazione;18 solo in pieno Settecento, infatti, una melodia emerge dalle carte di un processionale che fu stampato a Roma da Mainardi su iniziativa del Maestro generale Tomàs Ripoll (Processionarium, 1736, pp. 505-514) e ristampato dallo stesso editore per ordine del padre Antoine Bremond (1754, 499-508). Altre edizioni precedenti pubblicate da Giunta nel Cinquecento, e da Ciaccone-Paolini e Tinassi nel secolo successivo, sono prive di notazione in corrispondenza delle Passioni.19

È opportuno ricordare che il Gros livre, compilato nella seconda metà del Duecento, costituisce la più antica testimonianza di una divisione del testo della Passione fra

15. Marghera (Ve), Archivio della Provincia Veneta dei Frati minori, Romani cantus compendiolum, 1715; per una descrizione del manoscritto e della notazione cf. Toigo (2017, capitolo 1, paragrafi 2.1 e 5). La nota «ad cantores et lectores» si trova al fol. 1v. Figura 2 Marghera (Venezia), Archivio della Provincia veneta dei frati minori, 16. Cf. nota 2. Romani cantus compendiolum, 1715, carta 29v 17. I-Rss ms. XIV L 1, Ecclesiasticum officium, anno 1260 ca.

18. A dire il vero, un’intonazione domenicana intermedia esiste, ed è riportata alla fine di un’edizione giuntina (Missale, 1562); nemmeno questa melodia, però, può essere considerata di tradizione italiana, essendo stata approntata «iuxta decreta Capituli generalis, Salmanticae […] celebrati», cioè secondo i decreti di quel Capitolo che stabilì anche la revisione del messale e del breviario.

19. Processionarium Ordinis fratrum Praedicatorum, Venezia, Giunta, 1517; Processionarium secundum Ordinem fratrum 14. Padre Matteo Orlando (1668, Roma: Giacomo e Andrea Fei); padre Carlo Filiberto Barberi (1699, Roma: Gaetano Zenobi et Praedicatorum, Venezia, Giunta, 1517; Processionarium secundum ritum et morem fratrum Praedicatorum, Venezia, Giunta, Giorgio Placho); padre Luigi Laghi (1755, Venezia: Baglioni); padre Giuseppe Ximenez (1774, Venezia: Baglioni). Nell’edizione 1545; Processionarium iuxta ritum sacri Ordinis Praedicatorum, Roma, Ciaccone-Paolini, 1610 e 1619; Roma, Tinassi, 1661 del 1699, alle formule consuete vengono aggiunte le ultime parole di Cristo sulla croce nella versione di Luca (modulate sulla e 1679. Privi di formule notate sono anche il Processionarium sacri Ordinis Praedicatorum, Parigi, Convento di Saint Jacob, stessa di figura 1), insieme con un breve frammento dalla Passione secondo Giovanni. Purtroppo non v’è traccia di melodia nei 1707, il Missale ad usum fratrum Praedicatorum S. Dominici, Venezia, Giunta, 1550, e il ms. V-CVbav Vat. lat. 9214, un messali carmelitani del 1521 (Venezia: Giunta), del 1587 (Roma: Gardano-Coattino) e del 1769 (Venezia: Baglioni). Processionale-Rituale «monialium Ordinis Praedicatorum».

348 349 CANTUS PLANUS 2014 TOIGO — IL CANTO DEL PASSIO IN FONTI ITALIANE TARDIVE persone diverse.20 Le corde di recitazione sono rispettivamente il Do, il Fa e il Si Alla tradizione romano-francescana va ricondotta anche l’amplificazione melismatica bemolle, mentre le ultime parole di Cristo sulla croce, intonate in modo sillabico, che adorna tre momenti solenni della Passione nella sezione dell’Evangelista, ossia la rimangono nell’ambito melodico grave proprio del Cristo, al contrario di quanto cadenza che precede le ultime parole di Cristo sulla croce («clamavit Jesus voce magna accade nella tradizione italiana, dove in genere si ha un improvviso spostamento dicens»), le parole «tradidit (emisit) spiritum», e la conclusione del Passio («sedentes verso l’acuto. L’intera melodia si svolge secondo uno stile sobrio e sillabico, anche contra sepulchrum»: cf. esempio 4). nella parte dell’evangelium che segue la Passione. Dunque, i toni di recitazione del ‘prototipo’ domenicano non sono quelli della tradizione italiana, e non è sufficiente Esempio 4. Processionarium sacri Ordinis Praedicatorum, Roma, Mainardi, 1754, p. 507 una loro trasposizione per ottenere un’intonazione di tipo italiano. Tornando invece ai Processionali stampati dal Mainardi, le formule, chiaramente modulate sulle corde di Sol, Do, e Fa, presentano vari melismi e fioriture, e nei luoghi fondamentali richiamano con evidenza la tradizione romano francescana. Basti considerare l’initium del Cristo, dove si riconosce l’impronta del consueto melisma ascendente (esempio 1), oppure la melodia per le ultime parole di Cristo sulla croce, costruita sul doppio vocalizzo in Re (figura 3).21 L’intonazione contiene un elemento di originalità in corrispondenza delle grida della folla («Tolle, tolle, crugifige!», cf. esempio 5), dove la melodia sale improvvisamente al Si bemolle acuto, ben oltre la corda di recitazione che è propria delle Turbe. L’evangelium che segue il Passio, impropriamente introdotto dalla rubrica «finis Passionis», presenta una corda di recitazione che non coincide con quella del Gros livre, così come diversa è la finis evangelii («cum custodibus»), qui sottolineata da altra amplificazione melismatica.22 Dunque, il prototipo non ha esercitato alcuna influenza su questa intonazione tardiva, che a tratti risente del modello romano- francescano, e che evidentemente perdurò nella liturgia dell’Ordine, essendo attestata anche nel Processionarium sacri ordinis Praedicatorum stampato a Roma nel 1894 e nel Cantus Passionis uscito a Roma nel 1953.23

Esempio 5. Processionarium sacri Ordinis Praedicatorum, Roma, Mainardi, 1754, p. 507

Figura 3 Processionarium sacri Ordinis Praedicatorum, Roma, Mainardi, 1754, p. 502

22. Come afferma Rocha (2004, pp. 124-125), «mentre per la liturgia dei Frati Minori la fonte è chiara» (e cioè l’Ordinario della Curia romana), non è possibile stabilire a quali antigrafi abbiano attinto i Domenicani dei primi decenni nella 20. Come è noto, anche il Graduale Sarisburiense, risalente alla fine del Duecento e conservato a Parma (I-PAc ms. 98), redazione del prototipo. Quanto all’origine dell’Evangeliario contenuto nel Gros Livre, si consideri anche quanto scrive impone una divisione del testo fra cinque ‘attori’, essendo prescritta una prima vox per l’indirizzo iniziale del sacerdote Załuska ( 2004, p. 131): «Comme la grande majorité des Évangéliaires médiévaux produits au Nord des Alpes, en particulier («Dominus vobiscum») e una quinta vox per le ultime parole di Cristo sulla croce (Stäblein 1962, Tafel 61/2). en France, et contrairement à la vieille habitude romaine, l’Évangéliaire du prototype dominicain commence l’année liturgique avec le premier dimanche de l’Avent, usage qui paraît renouer avec l’ancienne coutume gallicane». 21. Vocalizzo da eseguire probabilmente senza il bemolle. Il Cantorino presenta anche delle formule alternative per le ultime parole di Cristo sulla croce, tutte costruite con frammenti melodici ricavati dalla prima melodia-Eli. 23. Processionarium 1894, pp. 424-429; Cantus Passionis 1953.

350 351 CANTUS PLANUS 2014 TOIGO — IL CANTO DEL PASSIO IN FONTI ITALIANE TARDIVE

L’intonazione olivetana Tornando al Direttorio, la sezione che comprende le formule di Passione è introdotta Ho già accennato al manoscritto ‘Illinois’, risalente agli inizi del sedicesimo secolo dalla seguente intestazione: «Metro alli rr. pp. hebdomadari, diaconi, sottodiaconi e riconducibile alla Congregatio Cassinensis di Santa Giustina.24 Un’altra fonte et accoliti» (p. 244). Le formule sono impaginate con ordine, a partire da p. 256 prodotta nell’ambito della stessa Congregazione, il Cantorino monastico del 1506,25 dove figura la «Voce dell’Evangelista»; seguono la «Voce di Iesu Christo», la «Voce contiene anche la melodia-Eli (assente in ‘Illinois’) secondo la formulazione romano- di Christo moribundo» (nella versione di Giovanni e di Matteo), la «Voce di Turba francescana con doppio vocalizzo; la melodia, d’altro canto, è attestata anche in altri sola», la «Voce di Turba in multitudine», la «Voce di Pilato» e la «Voce finale ambiti monastici, come dimostrano due messali a stampa usciti nel 1503, l’uno di dell’Evangelista». All’interno delle singole sezioni, alcune brevi indicazioni, poste origine vallombrosana (Missale monasticum 1503a) e l’altro di provenienza camaldolese a lato dei tetragrammi, precisano il tipo di cadenza richiesta.29 La notazione tende (Missale monastiicum 1503b). Agli inizi del Seicento risale la testimonianza di una a rileggere in chiave ritmica i tradizionali punti di forma quadrata e in forma di nuova intonazione monastica, riconducibile all’Ordine olivetano: fra gli anni 1615 e losanga, che alludono a figure di breves e di semibreves; si tratta di una tendenza 1616, infatti, Adriano Banchieri, monaco olivetano, curò l’edizione di un Direttorio originale, inaugurata dal celebre Directorium chori del Guidetti, al quale questo monastico, suddiviso in due libri di due parti ciascuno, che fu pubblicato a Bologna olivetano evidentemente si ispira anche nel titolo.30 dagli eredi di Giovanni Rossi.26 Nella dedica del primo libro, rivolta al Cardinale di S. Cecilia, vescovo di Albano e protettore dell’Ordine,27 l’autore attesta di aver composto Ferma restando la presenza dei toni ‘italiani’ (Sol/Do/Fa), le formule presentano il Direttorio «a requisitione del nostro reverendissimo Padre Generale, per pubblico alcuni elementi di originalità: la voce di Pilato, ad esempio, è separata dalla voce giovamento della studiosa gioventù olivetana». Il secondo libro è dedicato proprio al delle Turbe e intonata all’ottava inferiore, mentre il tono dell’Evangelista si abbassa priore, Clemente Cattanei di Bologna: a p. 204 il Banchieri ricorda ancora una volta di improvvisamente dopo il momento della morte di Gesù («tradidit spiritum»), aver ricevuto da lui l’invito «acciò reducessi insieme un Direttorio per uso particolare proseguendo in modo dimesso sulla corda di La; dunque, il cambiamento del tono olivetano e conforme al nuovo Breviario», e conclude omaggiando il Cattanei con la non avviene in corrispondenza dell’evangelium finale, dove è di norma richiesto, seguente squisita espressione: «Gradischi lei in tanto, con la sua solita allegra faccia, bensì nella «Voce finale dell’Evangelista», quella che chiude la Passione propriamente questa mia pronta volontà di onorarla». La seconda parte del primo libro si apre con detta (esempio 6);31 la formula di finis Passionis, anzi, ‘modula’ sulla nuova finalis una breve storia dell’Ordine olivetano, al termine della quale il Banchieri informa che Sol per mezzo di un Fa diesis (esempio 7). già il padre fondatore (l’abate Bernardo, al secolo Giovanni Tolomei) aveva introdotto nella Congregazione il «canto fermo alla romana»; a p. 119, Banchieri aggiunge anche Esempio 6. Direttorio monastico di canto fermo, Bologna, Eredi di Giovanni Rossi, 1615, p. 262 che tale canto «la Congregatione olivetana imita quasi al naturale, il che non succede nelle altre Congregationi di monaci, atteso che i metri loro sono, chi in parte e chi in tutto, differenti». Per quanto riguarda l’intonazione della Passione, in realtà, non v’è somiglianza con la tradizione romana, se per ‘tradizione romana’ intendiamo quella del Cantus Ecclesiasticus Passionis del Guidetti, che in quegli anni continuava a diffondersi grazie alle prime ristampe.28

Esempio 7. Direttorio monastico di canto fermo, Bologna, Eredi di Giovanni Rossi, 1615, p. 262 24. Cf. nota 13. Se in Rism B/IV/4, pp. 1172-1173 si attribuisce a ‘Illinois’ una provenienza francescana, Giulio Cattin (1972, p. 446, nota 4) ne ha rivelato l’appartenenza alla congregazione riformata dei Benedettini italiani. Quando, nel 1504, l’abbazia di Montecassino aderì alla riforma, la Congregazione assunse il nome di Congregatio Cassinensis, alias Sanctae Iustinae. In proposito: Leccisotti 1944; Sambin 1959; Tassi 1952; Trolese 1984 e Trolese 1988.

25. L’edizione, acefala, reca nel colophon la seguente iscrizione: «Monastici cantus compendiolum feliciter explicit. Venetiis per Lucantonium de Giuntis Florentinum, anno Domini 1506, pridie nonas maij». Le ristampe, pubblicate a Venezia, presso l’editore Giunta, nel 1523 e nel 1535, recano il titolo di Cantus monastici formula. L’intonazione cassinese ebbe evidentemente un certo peso, in ambito monastico, visto che la stessa melodia fu riprodotta in un Cantorino francese (Cantus monastici formula 1889, pp. 43-47).

26. Banchieri 1615. Ciascuna sezione del volume riporta un nuovo frontespizio; il Libro secondo e terza parte (p. 201) reca in calce la data 1616, che torna anche nel colophon dell’opera. Per questo importantissimo lavoro, cf. Torelli (2006, pp. 453-454). Quanto alla figura di Banchieri, cf. Mischiati (1972); May e Wiering (2001). Sulla storia della Congregazione, si veda Picasso (1975). 29. Per l’Evangelista: «Titoli», «Punti», «A Cristo», «Alla turba e Pilato», «Finale»; per il Cristo, le Turbe e Pilato: «Interrogativo», «Suspensivo». 27. In quegli anni effettivamente il titolo cardinalizio di Santa Cecilia era in mano a Paolo Emilio Sfondrati, vescovo di Albano. La dedica si trova a p. 4 del Direttorio. 30. Guidetti 1582. Sulle relazioni fra i due Direttòri, cf. Torelli (2006, pp. 448-450).

28. Cf. nota 8. 31. Nella tradizione cassinese la corda di La è riservata solo all’evangelium.

352 353 CANTUS PLANUS 2014 TOIGO — IL CANTO DEL PASSIO IN FONTI ITALIANE TARDIVE

Anche le formule olivetane rivelano una certa inclinazione all’espressività: un soluzioni semiografiche originali, utilizzando note bianche e nere che modificano melisma ascendente orna l’initium del Cristo; la cadenza dell’Evangelista che precede talune cadenze in senso più esplicitamente proporzionale (si veda la cadenza del le ultime parole di Cristo sulla croce si porta al registro acuto per introdurre la «Consumatum est» in figura 4). Questa tendenza semiografica ‘di transizione’, che melodia-Eli, come è tipico di molte Passioni italiane: la melodia-Eli, infine, anch’essa inserisce elementi ritmicamente connotati in un contesto notazionale ambiguo, si al registro acuto, sembra celare un’eco della formula romano-francescana, non solo ritrova ancora nel Cantorino settecentesco di Giovanni da Bassano (figura 2), ma nelle due invocazioni, ma anche nel tratto «lamasabathani?» (esempio 8). era già emersa, in ambito monastico, ben prima dell’opera del Guidetti: lo attestano le formule di Passione contenute in un Cantorino di origine monastica conservato Esempio 8. Direttorio monastico di canto fermo, Bologna, Eredi di Giovanni Rossi, 1615, pp. 257-258 a Bologna, il ms. 2931 della Biblioteca Universitaria,35 ma anche quelle della citata edizione monastica del 1506.36

Figura 4 L’intonazione del Direttorio monastico fu riproposta nel Cantorino utile a novizzi Cantorino utile a novizzi, Bologna, Eredi di Batolomeo Cochi per Bartolomeo Magni, 1622, p. 74 di Adriano Banchieri, stampato a Bologna nel 1622,32 e in un Cantorino olivetano approntato da padre Alfonso Bovio e pubblicato a Venezia nel 1661.33 Nella presentazione all’edizione del 1622 (p. 3) Banchieri, all’epoca abate di San Michele in Conclusione Bosco presso Bologna, informa di aver predisposto il Cantorino «per uso particolare Agli inizi del diciassettesimo secolo era già operante in Italia il modello ‘romano della nostra religione olivetana» e «per uso universale di qual si voglia giovinetto cinquecentesco’ di Guidetti, il cui Cantus ecclesiasticus Passionis voleva forse religioso principiante di canto fermo»; «l’officiare alla olivetana», egli dice, «è simile introdurre una dimensione più sobria e contenuta nell’esecuzione di questo recitativo alla romana», perché le regole sono comuni, come anche «le intuonationi Salmodie de liturgico. Tuttavia, le intonazioni tardive qui presentate testimoniano il persistere gl’Hinni, e pratica di leggere e conoscere le Chiavi, mutationi et Mano». Le formule di una certa propensione all’espressività drammatica: una propensione che attinge di Passione, ricavate da più evangelisti, sono riportate alle pp. 72-75.34 A indicare la spesso alle formule romano-francescane tardomedievali, le cui tracce inconfondibili corda di recitazione, Banchieri si serve di note bianche (breves caudatae o longae) continuano dunque ad affiorare, nel corso del diciassettesimo e diciottesimo secolo, seguite da custodes, quindi utilizza il punctum quadrato e quello in forma di losanga all’interno delle più diverse tradizioni liturgiche. per rimarcare l’andamento prosodico del testo. Tuttavia, il musicista introduce anche

32. Banchieri 1622.

33. Bodio, 1661. Nella pagina dedicatoria, il padre Alfonso Bovio dichiara di aver stampato il Cantorino essendo ormai introvabile il Direttorio monastico di Banchieri. Quanto alle formule di Passione, si rileva qualche errore di posizionamento delle chiavi a p. 111, nel quinto e nel sesto tetragramma; nella «Voce finale dell’Evangelista» la chiave di Fa e le note non sono allineate con le righe, specialmente nell’ultimo tetragramma, dove la corda di recita è ancora sul La, e non sul Sol, come potrebbe a prima vista sembrare. Purtroppo non è al momento reperibile una preziosa edizione del 1611, segnalata in Torelli (2006, p. 453, nota 10, Cantorino olivetano 1611).

34. Dopo un’ampia sezione dedicata a Matteo, si leggono le ultime parole di Cristo sulla croce secondo Giovanni, quindi un’altra sezione di Matteo con la voce di Pilato, un brevissimo inserto di Luca e la finis Passionissecondo Giovanni. Talvolta le espressioni sono abbreviate, come nel caso seguente: «Et circa horam nonam esclamavit dicens: Eloim, Heloi …» (Mt. 27, 35. I-Bu ms. 2931, Cantorino, secolo quindicesimo; cf. Stäblein (1962, col. 895, con trascrizioni alle cols. 891-894). Quanto 46; qui si nota una certa commistione con il testo di Marco); in altri casi il testo presenta delle varianti, come nell’espressione all’origine monastica del manufatto, cf. Cattin (1972, p. 460, nota 28). «Pasca fiet vobis» (Mt. 26, 2), o delle commistioni col testo di altri evangelisti, come nel tratto seguente: «Tunc dixit illi ancilla hostiaria: Et tu cum Iesu Galileo eras» (Mt. 26, 69). 36. Cf. nota 28.

354 355 CANTUS PLANUS 2014 TOIGO — IL CANTO DEL PASSIO IN FONTI ITALIANE TARDIVE

Bibliography Missale monasticum secundum consuetudinem Ordinis Vallisumbrose (1503a), Venezia: Giunta.

Banchieri Adriano ed. (1615), Direttorio monastico di canto fermo, per uso particolare della Congregatione Missale monasticum secundum Ordinem Camaldulensium (1503b), Venezia: Zanchi. olivetana […], Bologna: Eredi di Giovanni Rossi. Missale secundum ordinem fratrum Praedicatorum, iuxta decreta Capituli generalis, Salmantice, anno Domini – (1622), Cantorino utile a novizzi e chierici secolari e regolari principianti del canto fermo alla romana 1551 celebrati, reformatum et authoritate apostolica comprobatum […] (1562), Venezia: Giunta. […], Bologna: eredi di Bartolomeo Cochi per Bartolomeo Magni. Monastici cantus compendiolum (1506), Venezia: Giunta. Baroffio Giacomo - Antonelli Cristiana (1986), La Passione nella liturgia della Chiesa cattolica fino all’epoca di Johan Sebastian Bach in Elena Povellato ed., Ritorno a Bach. Dramma e ritualità della Picasso Giorgio (1975), Congregazione benedettina olivetana, in Dizionario degli Istituti di Perfezione, 1974- Passione, Venezia: Marsilio, pp. 11-33. 2003, 10 vols., Roma: Edizioni Paoline, vol. 2, pp. 1494-1496.

Bovio Alfonso ed. (1661) Prima [Seconda e Terza] Parte al Cantorino olivetano nuovamente stampato e Processionarium sacri Ordinis Praedicatorum (1736), Roma: Mainardi (ristampa 1754). registrato dalli libri chorali […], Venezia: Francesco Bodio. – (1894), Roma: Desclée Lefebvre. Boyle Leonard E. - Gy Pierre M. - Krupa Pawels eds. (2004), Aux origines de la liturgie dominicaine: le manuscrit Santa Sabina XIV L 1, Roma: École française de Rome (Collection de l’École française de Rocha S. J. Pedro (2004), Liturgia della cappella papale, liturgia dei frati minori e liturgia dei frati predicatori Rome, 327. Documents, Études et Répertoires, 67). in Boyle, Gy and Krupa 2004, pp.115-125.

Cantorino olivetano nuovamente stampato a richiesta del reverendissimo Padre D. Angiolo Maria Alchisio Roth Adalbert (1983), Zur Datierung der frühen Chorbücher der päpstilchen Kapelle in Ludwig Finscher Datierung und Filiation von Musikhandschriften der Josquin-Zeit Abbate Generale (1611), Milano: eredi di Pacifico da Ponte e G. Battista Piccaglia. ed., , Wolfenbüttel: Harrasowitz (Quellenstudien zur Musik der Renaissance 2, Wolfenbütteler Forschungen 26), pp. 239-268. Cantus monastici formula […] (1889), Tournay: Desclée-Lefebvre. Sambin Paolo (1959), Sulla riforma dell’Ordine Benedettino promossa da Santa Giustina di Padova in Paolo Sambin, Ricerche di storia monastica medievale, Padova: Antenore (Miscellanea erudita 9), pp. 69-122. Cantus Passionis d.n.j.c. secundum Matthaeum et secundum Joannem juxta notam Processionarii s.o.p. (1953), Roma: ad S. Sabinae. Sorbelli Albano ed. (1911), Cortona in Inventari dei manoscritti delle Biblioteche d’Italia, 1890-2013, 116 vols., Forlì: Bornandini, vol. 18 (ristampa 1966, Firenze: Olschki). Catena Claudio (1975), Carmelitani in Dizionario degli Istituti di Perfezione 1974-2003, 2, pp. 460-521.

Stäblein Bruno (1962), Die einstimmige lateinische Passion, in Mgg 10, cols. 887-898. Cattin Giulio (1972), ‘Canti polifonici del repertorio benedettino in uno sconosciuto «Liber quadragesimalis» e in altre fonti italiane dei secoli XV e XVI inc.’, Benedictina 19, pp. 445-537. Tassi Ildefonso o.s.b. (1952), Ludovico Barbo (1381-1443), Roma: Edizioni di Storia e Letteratura (Uomini e Dottrine 1). Compendium musices confectum ad faciliorem instructionem cantum chorale discentium necnon ad introductionem huius libelli: qui cantorinus intitolatur […] (1513), Venezia: Lucantonio Giunta. Toigo Diego (2011a), Dalla contemplatio alla compassio. Tendenze espressive in Passioni italiane fra Medioevo e Rinascimento in Cantus Planus: Papers read at the Fifteenth Meeting; Dobogókő, Hungary; Directorium chori una cum Processionali, iuxta ordinem ac ritum B. Mariae Virginis de Monte Carmeli, 23-29 August 2009, Budapest: Hungarian Academy of Sciences - Institute for Musicology, pp. 793-809. continens ea quae ad sacra officia cantu persolvenda pertinent[…] (1614), Napoli: Giovanni Giacomo Carlino. – (2011b), Intonazioni della Passione in area italiana nel sec. XVI. Il «modello romano» in Claudia Caramanna - Novella Macola - Laura Nazzi eds., Citazioni, modelli e tipologie nella produzione Dizionario degli Istituti di Perfezione (1974-2003), 10 vols., Roma: Edizioni Paoline. dell’opera d’arte. Atti delle giornate di Studio, Padova, 20-30 maggio 2008, Padova: Cleup (Scuola di dottorato in Storia e critica del beni artistici, musicali e dello spettacolo. Atti 2), pp. 249-254. Familiaris clericorum (1517), Venezia: Gregorio de Gregori. – (2012a), Una Passione di Cividale, Museo Archeologico Nazionale, ms. XXIV in Gozzi 2012, pp. 223-232. Gozzi Marco ed. (2012), Cantus fractus italiano: un’antologia, Hildesheim-Zürich-New York: Olms (Musica mensurabilis 4). – (2012b), Una Passione di Parma in Gozzi 2012, pp. 201-208.

Guidetti Giovan Domenico ed. (1582), Directorium chori ad usum sacrosanctae basilicae Vaticanae et – (2012c), Inedite fonti italiane della Passione in notazione ritmico-proporzionale in Cantus Planus: aliarum cathedralium et collegiatarum ecclesiarum […], Roma: Robert Granjon. Papers read at the sixteenth Meeting; Vienna, 21-27 August 2011, Vienna: Hungarian Academy of Sciences - Institute for Musicology, pp. 394-402. – (1586), Cantus ecclesiasticus Passionis, 3 vols., Roma: Alessandro Gardano. – (2015), ‘Fonti toscane tardive della Passione di Gesù Cristo’, Musica e Figura 3. Leccisotti Tommaso (1944), ‘La Congregazione benedettina di Santa Giustina e la riforma della Chiesa al secolo XV’, Archivio della Società Romana di Storia Patria 67, pp. 451-469. – (2017), Intonazioni monodiche della Passione in Italia fra i secoli XIII e XVI, Padova: Cleup (Fonti e studi per la storia della musica veneta 5). May S. William - Wiering Frans (2001), Adriano Banchieri in Grove7, pp. 619-621. Torelli Daniele (2006), Notazioni ritmiche e canto fratto nelle edizioni non liturgiche tra Cinquecento e Mischiati Oscar (1972), ‘Adriano Banchieri (1568-1634). Profilo biografico e bibliografia delle opere’, Seicento in Marco Gozzi - Francesco Luisi eds., Il canto fratto. L’altro gregoriano, Atti del Convegno, Annuario del Conservatorio di Musica “G.B. Martini” di Bologna 1, pp. 37-201. Parma - Arezzo, 3-6 dicembre 2003, Roma: Torre d’Orfeo (Miscellanea musicologica 7), pp. 447-492.

356 357 CANTUS PLANUS 2014

Jean-François Goudesenne Trolese Francesco G.B. ed. (1984), Riforma della Chiesa, cultura e spiritualità nel Quattrocento veneto. Atti del Convegno per il VI centenario della nascita di Ludovico Barbo (1382-1443), Padova-Venezia-Treviso, 19-24 settembre 1982, Cesena: Badia di Santa Maria del Monte (Italia benedettina 6). Apostolicité et exports liturgiques en France méridionale,

Trolese Francesco G.B. (1988), Decadenza e rinascita dei monasteri veneti nel basso Medioevo in Francesco Gaule cisalpine et Italie non romaine (835-1150) G. B. Trolese ed., Il monachesimo nel Veneto medioevale, Atti del Convegno di studi in occasione del Millenario di fondazione dell’abbazia di S. Maria di Mogliano Veneto, Treviso, 30 novembre 1996, Cesena: Badia di Santa Maria del Monte, pp. 169-199.

ZaŁuska Yolanta (2004), Évangéliaire du ‘prototype’ dominicain et Évangéliaire du ‘prototype’ cistercien in Boyle, Gy and Krupa 2004, pp. 127-157. La question de l’apostolicité dans le développement et l’accroissement du sanctoral des antiphonaires des églises du monde latin traverse l’histoire de l’office depuis les strates les plus centrales des septième-huitième siècles, jusqu’aux nouvelles com- positions qui ont fleuri à la réforme grégorienne et qui ont vu se multiplier de nom- breuses historiae. Leurs modèles de sainteté allaient chercher des personnages qui, au départ, s’inscrivaient dans le prolongement de l’église apostolique; progressive- ment elles sont allées puiser dans une hagiologie toujours plus proche, directement ou indirectement, de la figure du Christ. En marge du vieux-fonds romain des offices des apôtres Pierre, Paul, Jean l’évangéliste, quelques grands monastères francs ont joué à cet égard un rôle important aux huitième et surtout neuvième siècle. Nous pouvons déceler une influence notoire de l’apostolicité de saint Denis en Aquitaine puis en Provence et surtout en haute Italie, à Verceil et Pavie. Par ricochet, à Venise où comme l’a montré Giulio Cattin, le culte de saint Marc effectue un large emprunt à celui de saint Syr, évêque de Pavie. Plus tard, au moment de l’émancipation des comtés et des évêchés, des centres peu connus, tissent des réseaux de diffusion pro- bablement liés aux transferts de reliques, notamment entre Lipari et Bénévent, puis de Bénévent vers le Nord jusqu’au Brabant et en Frise: ils continuent, à l’instar d’Hil- duin avec saint Denis (Lapidge 2017), d’Adhémar de Chabannes avec saint Martial (Grier 2018), à ériger l’apostolicité comme une revendication hautement réclamée au moment de la réorganisation administrative des églises locales. D’où cette florai- son d’offices en Provence, qui oppose avec Trophime les évêchés d’Arles et la puis- sante abbaye Saint-Victor de Marseille, adossée à la cathédrale de la Major; autour d’Aoste avec Simon et Jude; à Novare, un office inédit pour saint Matthieu traduit également des intentions apostoliques, érigées pour asseoir les prétentions d’admi- nistration d’églises comme Gozzano, sur les rives du lac d’Orta, à un niveau diocé- sain. Initialement liés à des saints fondateurs artificiellement rattachés à la période apostolique par une littérature hagiographique pseudo-apostolique qui caractérise la période sise entre la fin du huitième siècle et 900, des offices pseudo-apostoliques vont fleurir peu avant la réforme grégorienne, à l’instar de sainte Marie-Madeleine ou du cycle provençal des Maries, dont le retentissement sera assez conséquent dans la chrétienté latine (Sachser 1949).

Il nous semble important dans ce contexte historique, de procéder à une analyse cri- tique des formes et structures musicales, afin d’établir une chronologie plus fine des procédés de composition, notamment pour montrer comment se sont élaborées, très

358 359 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150) progressivement, les techniques de contrefaçon entre le neuvième et le onzième siècle Le plus ancien office tourne autour de la figure de saint Barthélémy, dont les reliques et distinguer certains timbres ou caractéristiques mélodiques comme représentatives ont été apportées de Lipari à Bénévent au neuvième siècle (Bozóky 2006); on retrouve d’une période précise. sa diffusion dans plusieurs autres centres, à Asti puis à Saint-Thierry de Reims, pro- bablement par les translations ou exports de reliques. L’analyse des textes comme des Le paysage historique du culte liturgique des saints ‘apostoliques’ mélodies, va faire émerger l’importance du modèle dionysien, fortement promu par Sans remonter aux offices apostoliques les plus centraux du répertoire romain, autour Hilduin de Saint-Denis dans les années 830 (Goudesenne 2002a; Lapidge 2017). Dans des apôtres Jean, Paul et Pierre, qui constituent le socle du ‘vieux-fonds’ romain et les foyers de haute Italie, autour de Pavie, le saint patron de cette cité dont le culte s’est romano-franc, nous remarquons à travers le sanctoral des zones que nous avons ob- diffusé dans quelques centres voisins, revêt une certaine importance dans la mesure servé avec la Provence, les Alpes et la grande plaine Padane, d’Ivrée jusqu’à Venise, en où il constitue le modèle de l’office de saint Marc, très emblématique sur le plan his- passant par Pavie et Vérone, quelques corpus liturgico-musicaux qui viennent revêtir torique dans la création d’un culte apostolique opéré par translation/vol de reliques une importance notoire. spécifiquement destiné à l’affirmation politique et ecclésiastique de la cité des doges (Cattin 1990-1992). Nous pouvons donc distinguer d’emblée deux phases dans la typologie des offices apostoliques: une première phase qui s’inscrit dans le prolongement de la littérature Tableau 2. Seconde phase (950-1050) hagiographique légendaire des huitième et neuvième siècles, liée aux nécessités d’une narration historique de l’évangélisation, notamment des Gaules (tableau 1). Elle ca- Ordre modes Centres promoteurs Modèles Influence et diffusion ractérise un corpus qui s’est élaboré dans le cours du neuvième siècle et développé régulier rythmique jusqu’au milieu du dixième siècle, très proche de la typologie des martyrs et souvent lié au récit légendaire des douze apôtres des Gaules, inspiré de Grégoire de Tours.1 Martial Limoges, Adhémar Denis? Cambrai (1029-1031) Anne × × Nombreux centres Tableau 1. Première phase (800-950) Jacques A B C × × Mogont Germanie Ordre modes Centres promoteurs Modèles Influence, Vézelay/Compostelle Europe centrale zone d’expansion Lazare × Marseille régulier rythmique Marie-Magd. A B C × Ste-Baume, Aix, Vézelay Conques, Fleury, 2 André A B, compl. × Francie nord-ouest, nombreux centres 3 Hucbald? Maries Jacobé × × Provence Reims, Mayence Apollinaire Ravenne Ivrea et Salomé Marseille? Cracovie Marthe A × Provence Barthélémy Bénévent Denis Reims, Pays-Bas Marthe B × × Tarascon Provence Denis A – B St-Denis, Hilduin, ca. 835 Haute Italie Saturnin Matthieu Novare-Gozzano Marc Venise Syr Maximin Ste-Baume, Aix Pierre (Chaire) A B × Reims (Hucbald) Italie Simon & Jude × × Aoste, Alpes, Gozzano Plusieurs centres épars Remi Denis Fillastre (Brescia) Trophime Arles Marseille Saturnin Saturnin Toulouse Denis Provence Catalogne 2 3

La seconde phase (tableau 2) correspond davantage à une phase d’un renouveau des 1. Sur les légendes apostoliques de la fondation des Gaules d’après Grégoire de Tours, voir Zeiller (1926). cultes et à une ecclésiologie nouvelle, étroitement liée à la réforme grégorienne, qui se 2. Le terme Francie (Francia en latin) est plus approprié pour l’époque carolingienne que celui de la France, qui n’est manifeste à partir du milieu du onzième siècle, à l’exemple du mouvement insufflé par qu’effectif après Hugues Capet, voire le douzième siècle. Elle se divise encore aux neuvième-dixième siècles en Francie occidentale, médiane (de la Lorraine à Bénévent) et orientale (Germanie). Adhémar de Chabannes dans la promotion du culte de saint Martial (Andrault-Schmitt 3. Jacobsson 2008, p. 173. 2006). En Provence, cette quête de l’apostolicité, thème central de sainteté dès l’époque

360 361 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

carolingienne, s’est cristallisée autour de saint Trophime d’Arles, sur le modèle de Autres lieux Provence Francie occidentale Alpes Padanie Rome l’office de saint Saturnin, lui-même rattaché à l’office de saint Denis, l’évêque de Paris confondu avec l’aréopage, disciple de saint Paul (Goudesenne 2004). Arles se focalise 900 Chaire de saint RAVENNE Pierre (Hucbald) Apollinaire alors sur Trophime, dont elle use de l’apostolicité pour des revendications qui ont été dixième METZ: Etienne inv. Victor fabriquées très probablement dans le cours du onzième siècle, comme la plupart des siècle offices locaux des autres diocèses(Goudesenne 2016). Des antécédents ont précédé An mille Benoît, transl. Trophime NOVAR E ce ‘cycle’ avec des saints apostoliques ou des fondateurs, Trophime et Victor, entre le Honorat Matthieu cours du dixième siècle et 1050 (Duprat 1943-1946). Le contexte féodal des comtés et 1030 Martial (Adhémar) la diffusion des reliques, d’une large provenance, n’enferme pas ces diocèses, mais les expose aux vents d’une histoire tumultueuse et brosse un panorama singulier et 1040-1050 VÉZELAY: Marie-Magd. Simon prolifique de ces terres provençales parmi les paysages liturgiques de l’Europe latine Marie-Magd. et Jude de la fin du haut Moyen Âge (Goudesenne 2002b). S’y agrègent de nombreux offices de CONQUES: Foy Jacques; Nicolas saints ‘nouveaux’, à l’exemple de Simon et Jude dans les Alpes piémontaises, ou encore avec de nombreux saints ‘anciens’ puisés dans l’entourage du Christ, dont les offices 1100 CLUNY: Marthe A Transfiguration Maximin sont composés selon des critères techniques et esthétiques nouveaux par rapport aux strates précédentes (Jacobsson-Haug). À partir du second millénaire, dans le cours 1144 Anne? Marguerite? Maries Jacobé Augustin? et Salomé du onzième siècle, c’est surtout la translation de Madeleine de la Sainte-Baume vers 1187 la Bourgogne qui va constituer le corpus prégnant d’une sainteté peut-être pas stric- Thomas Becket Marthe B Lazare tement apostolique, mais qui en tous cas, recherche la promotion du culte de témoins Après 1200 Conception, Maxime de directs du Christ et de l’église apostolique: Marie-Madeleine, Marie Jacobé, Marie Visitation Riez? Salomé, Marthe, Lazare, etc. (Duprat 1943-1946). Il est indéniable que les moines de Dominique Saint-Victor et les chanoines de la Major ont joué un rôle de premier plan dans l’éla- boration d’un corpus local à vocation régionale, que représente, entre 1040 et 1140, le Cycle des saintes femmes, incluant les offices de sainte Marie-Madeleine, Marthe,4 Lazare de Béthanie et les Maries Jacobé et Salomé. Cette question de l’apostolicité, au L’analyse musicale pour différencier cœur d’une véritable identité régionale aux onzième et douzième siècle fut construite à les confections d’une pseudo-apostolicité partir d’éléments divers, notamment un office d’origine bourguignonne (Saxer 1949a, 1949b, 1955, 1958, 1959; Saxer et Celletti 1967; Lobrichon 1989; voir tableau 3). Des ‘faux’ hagiographiques et diplomatiques pour justifier les revendications apostoliques. Le vecteur de l’apostolicité traverse les siècles et ne relève pas Tableau 3 . Offices étudiés et chronologie forcément d’une période spécifique, mais il revient avec davantage d’acuité dans le

Autres lieux Provence Francie occidentale Alpes Padanie Rome second millénaire, dans un contexte de confection de nombreux corpus liturgiques, eux-mêmes issus de dossiers hagiographiques fabriqués de toutes pièces, à l’instar des septième- Paul huitième Pierre faux diplomatiques, à l’exemple des cultes de saint Benoît et de saint Maur sous Désiré siecle au Mont Cassin (Goudesenne 2012). Il importe alors de recourir à la musicologie, 835 Denis (Hilduin) particulièrement à la critique détaillée des mélodies et l’examen des modèles, des Remi (Hincmar) influences, pour distinguer ces historiae apostoliques. Pour établir celles qui ont pu VENISE: Marc constituer un modèle influent, repérer les contrefaçons les plus flagrantes, distinguer fin du TOULOUSE: Saturnin Compléments à l’intérieur d’un office les parties les plus anciennes des additions ou compléments neuvième BÉNÉVENT: pour saint André composés à une autre époque. Des données essentielles pour d’autres disciplines, la siècle Barthélémy philologie et l’hagiographie, qui nous apporteront également d’autres critères pour

4. Le choix de la date du 29 juillet, octave de Marie-Madeleine n’est pas hasardeux et démontre l’intention de célébrer cette une meilleure contextualisation historique de ces compositions qu’on peut situer entre fête dans la continuité de celle de Madeleine, le 22; c’est d’ailleurs à cette date qu’aurait été instaurée la foire de Beaucaire, 800 et le milieu du douzième siècle pour les plus tardives. voir Olivier (2010, p. 40).

362 363 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

Évidence des modèles francs en Italie ‘non romaine’. L’ historia la plus importante Exemple 1 1. Barthélémy (I-ASs, ms XXIX, fol. 134v) qui appartient à cette typologie, probablement modelée sur des offices Francs en 2. Marc (Venise, éd. Cattin) lien avec les milieux de la cour Palatine, est saint Syr, fondateur du diocèse de Pavie, 3. Barthélémy (I-ASs, ms XXIX, fol. 124v) 4. Barthélémy (I-BV 21, 185v) patron du Tessin et de la Lombardie (Ogliari 2007). Il reçut un culte important qui se (1, 3 et 4 ont la même mélodie 1) développa autour de Pavie jusqu’à toute la Lombardie et même dans certaines parties du Piémont, constituant ainsi un des offices les plus diffusés de haute Italie – la plupart des offices des saints patrons étant souvent locaux et tout au plus diocésains. Toutes les antiennes de cette historia reçoivent un verset propre pour la récitation psalmodique, extrait non pas des psaumes, mais des textes hagiographiques. Comme les historiae de saint Denis ou de saint Remi, le choix de cette typologie d’office ‘basilical’ n’est évidemment pas un choix musical mais plutôt idéologique. L’apostolicité avérée du saint et son association spirituelle avec saint Pierre, ces revendications apostoliques relèvent probablement des souhaits de quelques prélats, de maintenir son clergé dans une grande proximité avec le pouvoir politique ducal ou royal. Dans l’office donc, les antiennes à verset, l’appartenance modale et les hypotextes s’apparentent à ces offices basilicaux (Denis, Paul, Laurent), mais la présence de quelques contrafacta et la trace de centonisation, de prosulation et d’adaptation, infirme l’appartenance de ce corpus à Exemple 2. Le neume triple du verset pour Saint Barthélémy la période précédente, fortement modelé par saint Denis, peut-être par l’intermédiaire de l’office en prose de saint Eusèbe de Vercelli (Goudesenne 2009; Vitale forthcoming). Il s’agit donc plutôt de répertoires élaborés plus tardivement, dans la seconde moitié du dixième siècle voire un peu plus tard encore.5

Modélisation sur les antiennes «O» de l’Avent6. De nombreuses antiennes, particulièrement pour les heures, reprennent ce timbre très connu des antiennes «O» de l’avent, connu pour être un apport franc au corpus romain (Claire 1986; Gasser 1992). Un timbre ancien qui se prête donc aisément à une modélisation sur des textes composés sur le modèle d’oraisons, qu’on retrouve pour de nombreux types l’office de saint Trophime d’Arles, à la même place, c’est-à-dire l’ouverture du cy- de saints martyrs ou confesseurs. La prégnance des modèles francs avec saint Denis cle de répons, mais encore pour des offices plus tardifs comme saint Barthélémy à ou saint Martin, se manifeste dans de nombreuses fêtes patronales de la péninsule Bénévent et Asti ou l’évangéliste Matthieu à Novare et Gozzano. (Barthélémy, Marc, Syr, Zénon de Vérone, Eusèbe de Verceil, Fillastre de Brescia, Prosper, etc.; voir exemple 1, p. 365). Au passage, c’est dans ce groupe de répons que l’on retrouve le célèbre neume triple décrit par Amalaire (Kelly 1988, Hanssens 1948-1950), qui ouvre le verset pour saint Au niveau des répons, on constate également la prégnance du premier groupe du Barthélémy (exemple 2). protus, qui en général ouvre souvent le cycle des historiae, sur le modèle du premier répons de l’Ascension Post passionem, du même type que le Répons Descendit de Autre groupe fréquent, celui du tétrardus, qu’on retrouve dans maints offices en celis de Noël. Les exemples suivants illustrent ce topos hagiographique qui situe les prose et qui, à Venise, pour l’office de Marc, s’accommode de l’hexamètre (Cattin apôtres-martyrs dans la continuité de la passion du Christ: Saturnin de Toulouse, 1990-1992, 3, p. 260). Les extensions au tétracorde supérieur (Re-Sol) caractérisent dont saint Denis représente un des modèles probables, et que l’on retrouve dans une phase plus tardive de la composition, notamment dans cet office local de saint Matthieu à Novare et Gozzano, près du lac d’Orta. Remarquons les approximations

5. La datation de tels corpus ne peut reposer sur la seule étude stylistique des mélodies; l’évolution du style du cantus n’est rythmiques des formules-types des versets dans les témoins, par exemple à Venise pas homogène et tout dépend des modèles, qui parfois ont pu être effectués tardivement malgré l’aspect ancien du style ou à Novare, où les offices de saint Marc comme de saint Matthieu semblent corres- (prose et formules-types du vieux-fonds). pondre à une couche plus tardive que celui de saint Barthélémy à Bénévent, qui a 6. Pour Barthélémy, transcription d’après l’antiphonaire Asti, Seminario vescovile, I-ASs ms. xxix, douzième-trosième siècle, voir Goudesenne (2014, pp. 6 passim). gardé quant à lui une écriture ornementale rythmiquement plus précise.

364 365 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

Exemple 3 Le contrefait commence avec ces pièces génériques à l’usage polyvalent pour un pe- Felix namque es almifice pontifex SirusC ao 6724 1. I-Novasd A1, fol. 74 tit groupe de saints, constituant un pseudo-commun des apôtres, le plus bel exemple 2. GB-Wo 160, fol. 243v, fac-similé: PalMus 12, p. 357. étant celui du réponse Felix namque, dont les premières apparitions dans l’office 3. F-As 465, fol. 439v se retrouvent à la fin du dixième siècle pour les fêtes mariales, aux côté du réponse Agmina sacra, affecté d’abord aux pontifes Germain et Vaast, présents dans l’anti- phonaire de Compiègne.7

R. Felix namque es, almifice pontifex Syrus, et omni laude dignus, *Qui extremos ad occasus Italiae fines pretiosae fidei gemmis primus condecorasti. V. Ora pro populo, interveni pro clero, et pro plebe postula (Cao 6724; voir exemple 3, p. 366).

Phase préliminaire et oralité relative du contrafactum. À la différence du type de contrefait ‘falsifié’, voici quelques exemples qui montrent que la modélisation s’est effectuée de façon transitoire et que la technique du contrefait s’élabore progressivement. Par exemple, voici un texte relativement courant qui pourrait s’apparenter au commun des confesseurs O beate N magna es fides tua (Cao 3999), qu’on retrouve pour Barthélémy avec une variante phonétique (caritate pour qualitate) et surtout, avec une mélodie-type du tétrardus (huitième mode), au lieu de la mélodie du protus donnée pour l’office de saint Denis. Une mélodie qui n’a pas été homogénéïsée par l’octoechos et qui se termine en Re sur La (avec différentia du deuxième mode). Des erreurs de transmission qui caractérise à cette phase une part d’oralité encore importante au niveau de la modélisation.

Le contrafactum intégral par substitution ou le ‘faux’. Toute la série des premières vêpres de la liturgie de Saint Marc à Venise, comme l’avait remarqué Giulio Cattin (1990-1992, 3, pp. 110 et 18*), reprend l’office de saint Syr de Pavie. Plusieurs caractéristiques montrent que l’original est bien ce confesseur martyr et non l’apôtre dont l’église des doges préfigure les cultes résultant d’un vol de reliques, comme saint Nicolas à Bari. C’est le Tessin, qui s’était enorgueilli des reliques du martyr, ici remplacé par Venise.

O quam beata urbis habitacio Ticinensis, in qua gloriosa menbra Syri in pace sepulta sunt, cuius Le contrefait. Pourtant, la question de l’apostolicité se recoupe avec des questions anima paradisi possidet aulam promisso sibi grege exorare non cessat. d’esthétique dans la composition musicale, au sens où les modèles de sainteté dans V. In qua eius florent oraciones memoria eius […]. ce ‘modèle apostolique’, au centre de très importantes revendications identitaires et ecclésiologiques, favorisent l’émergence de modèles et de compositions contrefaites, Marc (éd. Cattin): qui suivent plusieurs types d’élaboration. On aboutit ici à la distinction d’au moins O quam beata urbis habitacio Venetie, in qua gloriosa menbra Marci in pace sepulta sunt, cuius quatre types de contrafacta, qui nous invitent à opérer une très grande différenciation anima paradisi possidet aulam promisso sibi grege exorare non cessat. et d’importantes nuances dans la question des emprunts et de l’élaboration de ‘faux’, qui est aussi complexe et pas simplement une copie ‘mécanique’ par simple emprunt avec substitution du patronyme! 7. F-Pnm lat. 17436, ce répons n’y figure pas encore.

366 367 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

L’antienne Gratia superna V. Omnium quidem insiste sur la transmission doctrinale Dans les trois prosules composées pour la fin des trois nocturnes de matines, pour les- de la chrétienté revendiquée par Syr, qui en reçut l’autorité par son père spirituel, quelles on ne conserve pas, hélas, de mélodie dans les rares livres d’Arles, Trophime est l’antiochien Hermagoras, lui-même disciple de saint Marc et donc de saint Pierre. présenté comme un «prince des Gaules, qu’il évangélisa de son verbe, disciple asiate du L’ historia de saint Syr illustre pourtant une couche plus tardive du chant grégorien, maître Paul, docteur des païens»; il figure dans une sorte de triptyque aux côtés de Pierre probablement plus récente que la romanisation carolingienne; néanmoins, l’adoption et de Paul, porté par ce dernier répons s’inscrit dans cette apologie d’un «siège pontifi- de particularités esthétiques relève d’abord des points de vues idéologiques cachés cal placé sous la protection directe de ces deux fondateurs de l’Eglise», topos que l’on derrière ce culte, plus que de la musique à proprement parler. Les compilateurs litur- retrouve dans l’iconographie de la cathédrale (Hartmann-Virnich 1999a; 1999b; 2015). giques vénètes n’ont probablement ignoré la filiation apostolique du patron de Pavie, avec Hermagoras, qui a pu être mise à profit de saint Marc et justifier un tel emprunt. Outre ces trois longues prosules dont on ne conserve que le texte, sans mélodie, et qui indiquent le cours du onzième siècle pour leur composition, voici la réécriture du Syr (Ivrea, I-Iv 106; Novara, Archivio diocesano, I-Novasd A 1): répons Perfectum Christo famulum (CID 601787), qui traduit la volonté d’une différen-

Gratia superna per beatorum pectora Tycinensi populo percucurrit, ita a Christo in beatum Petrum ciation entre les deux centres, malgré le patronage plus marqué pour la cité d’Arles 9 a Petro in Marco vero in Armagoram, cuius beati patres nostri Syrus atque Yventius auditores deo dans les textes (exemple 4, p. 370). digni estiterunt. V. Omnium quidem precibus adiuvari petimus, quorum distributione sacram doctrinam suscepimus. S’il est difficile de porter un avis définitif sur cette réécriture mélodique qui distingue ces deux versions, celle d’Arles semblerait plus récente, se déployant dans une moda- Contrafactum avec concurrences entre cités épiscopales: Arles, Marseille et lité plus simple, moins ornée, moins aboutie et simplifiée. L’ornementation abondante saint Trophime. Même s’il est probablement postérieur à l’office de saint Victor, de l’antiphonaire de Marseille, plus riche que le bréviaire arlésien dans ses signes l’office de saint Trophime est sans conteste le plus intéressant du corpus: il est un graphiques quant à la notation musicale, se retrouve dans les autres répons et pour- des rares à faire l’objet de compositions réécrites et de compléments qui traduisent rait peut-être indiquer quelque influence des traditions hispaniques, car rares sont les à l’évidence une sorte de compétition entre Arles et Marseille. L’église d’Arles en a offices du onzième siècle qui portent un tel style de floraison ornementale. fait son patron quasiment exclusif, fermée à tout autre import de nouvel office des diocèses voisins, y compris Honorat et Césaire qui pourtant auraient pu honorer Compositions locales et modélisation problématiques cette antique cité - ils sont célébrés à Lérins et à Digne au plus tard au quinzième Parmi les historiae du deuxième groupe (du second millénaire), figurent incontes- siècle (Mathon 1967). Avec cette chapelle dédiée à Trophime, dont le tombeau fait tablement Marie-Madeleine et Lazare, qui vont être les principaux constituants du l’objet d’une vénération spécifique dans la liturgie, par exemple à Noël ou aux Saints- Cycle provençal dont l’apogée se situe entre 1050 et 1150 (Morin 1909, pp. 24 passim, Innocents, on mesure l’importance de ce culte central en Arles (Lorequais 1934, 2, Goudesenne 2016). Si l’ordre régulier des modes et le profil des timbres d’antiennes n. 456, pp. 431-434, fol. 82). L’enjeu est de taille car on en fait un disciple de saint comme des répons ne pose pas de doutes quant à la facture plus tardive des pièces, voici Paul, qui mérite même d’être hissé au panthéon national comme évangélisateur de des aspects problématiques dans la modélisation: surtout pour Lazare dont le style du l’ensemble de la Gaule, ce que rapportent les antiennes, et répons, enrichis par ces répons Occurrerunt Maria et Martha, très prolixe, tranche avec la brièveté condensée prosules d’un bréviaire-ordinaire du milieu du quatorzième siècle (F-Pnm lat. 752, de certains offices rythmiques, tels celui de Simon et Jude dans les Alpes. L’écriture du fol. 82-83) hélas sans musique.8 verset, très long, n’est pas modelée ni sur les formes classiques du grégorien, ni sur les nouvelles tournures des offices de cette époque. Comme le répons précédent de saint Trophime, plusieurs indices d’un remploi de pièces renvoyant à la liturgie ambrosienne 8. Leroquais 1934, p. 431-432, neuvième répons, avec prosule. Perfectum Christo famulum Petrus et Paulus Trophimum perficiunt in catedra pontificali gracia, quem dirigavit ad Gallias ou à des traditions parallèles caractérisant Saint-Victor de Marseille, viennent brouiller et nationes barbaras *Quo fidem Christi predicet ecclesias que fabricet. V. Primo quidem Arelato predicavit populum huic ad plures suos gentes dirigit discipulos V. Gloria Patri […] la simplicité supposée de la composition des nouveaux offices. Le répons pour Marie- postea repetitur r‹esponsorium> quo completo dicitur prosule Madeleine O felix sacrorum est également plutôt rare (figure 1, p. 372-373).10 Facta Christi predicans pco veritatis vir aplicus luciflua gentibus semina V. Sicut corda turgida Trophim rigando prophanis ritibus obiciens invida gmina V. Quippe qui ab ipero Petro est clcs ac Paulo principe minime piguit Ihu precepto illum iacendo sparge dogmata 9. Voici quelques incidences de la mention d’Arles dans l’antienne des Premières vêpres: «Galliarum principatus Arelati V. Olim asiana stirpe prole clara satis prosapia protoagoniste te caritatis nec ne Nichodemi nat matrera est collatus ob honorem et ornatus Trophimi apostolatus […]»; dans l’antienne du second nocturne «Nempe Galliarum V. Hic competit propria subactnativa patria Trophimus dignus […] unde Arelas […]», etc. V. Pauli carus a seculo factus affada famina V. Petri notus presentia 10. Il en est de même pour certains répertoires italiques, tel des répons de saint Apollinaire de Ravenne dans un fragment V. Disponitur a dextera d’antiphonaire conservé à la Bibliothèque capitulaire d’Ivrea (I-IV), dont l’écriture des versets n’a rien de standard avec la V. Quo inimico diluat phana et deo dicanda *fabri tradition grégorienne courante, voir Goudesenne (2009, pp. 17 et 25).

368 369 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

Exemple 4. Deux versions du répons final de l’Historia Trophimi (transcription de l’auteur) On ne peut donc pas vraiment parler d’une typologie spécifique d’office apostolique, 1. Marseille (F-Pnm lat. 1090, fol. 133v) 2. Arles (F-Pnm lat. 1091, fol. 11) mais constater des processus de réécriture fréquents tout au long de l’élaboration des répertoires de l’office entre 750 et 850, puis de façon accrue lors des réaménagements après l’an mille, avec les apports de nouveaux offices vers 1100-1150. Cette étude ré- vèle des circuits de diffusion liés aux translations de reliques, qui jusqu’à présent, ne semblaient pas avoir été remarqués dans les travaux critiques d’édition de l’of- fice, notamment cette tradition indépendante de Saint-Victor de Marseille découverte vers 1965 par le chanoine Lemarié, dont les ramifications vont jusqu’aux marches de Catalogne, à Ripoll, en Aquitaine méridionale, en Languedoc et au travers de ce qu’on appelle aujourd’hui la Provence (Lemarié 1965). Mais voici encore d’autres centres et réseaux en Gaule Cisalpine et dans cette vaste pianura padana, particulièrement remarquable dans la combinaison qu’elle opère entre les traditions locales et les di- verses influences, franques de l’ouest, ottoniennes et ambrosiennes.

370 371 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

Figure 1. Historia Mariae-Magdalenae et Historia Lazari, episcopi Massiliensis (F-Pnm lat. 1090, fols. 185v et 251v)

372 373 CANTUS PLANUS 2014 GOUDESENNE — APOSTOLICITÉ ET EXPORTS LITURGIQUES (835-1150)

Appendix – (2002b), ‘Les Offices des saints patrons provençaux: apports de la musicologie à l’histoire ecclésiastique d’Arles et de Marseille (Xe-XIVe s.)’, Les cahiers de Fanjeaux 37, pp. 113-146.

Chants cités – (2004), ‘La propagande aréopagitique dans la musique de l’office de saint Denis (IXe-XIe s.)’, Agmina sacra angelorum V. Potuit transgredi, répons Hagiographica 11, pp. 81-112. Egregius Christi Petrus V. In dei igitur Romam, Marc, antienne Felix namque es V. Ora pro populo, Marc, répons – (2009), Historiae from Alta Italia and their Frankish Models: a Progressive «Romanisation» by the Gratia superna V. Omnium quidem, Syr, antienne Carolingian Franks (750-950) in Roman Hankeln ed., Political plainchant? Music, text and historical In normes populi V. Exclamante universo, Saturnin, répons context of Medieval Saints’ Offices, Ottawa: The Institute of Mediaeval Music (Historiae Series, O admirabile gaudium, Barthélémy, antienne 2e vêpres Musicological Studies 111), pp. 13-29. O beate Bartholomee magna es fides tua, Barthélémy, antienne O felix sacrorum V. Angelica pollet, Marie-Madeleine, répons – (2012), Montecassino-Glanfeuil-Paris: circulation et différenciation d’un corpus romano-bénédictin aux O quam beata urbis, Marc, Syr, antienne IXe-Xe s.: l’office de s. Maur in Nicola Tangari ed., Musica e liturgia a Montecassino nel medioevo, O quam preclarus apostolus, Barthélémy, antienne Roma: Viella, pp. 199-228. Occurrerunt Maria et Martha V. Videns Iesus, Lazare, répons Perfectum Christo famulum V. Primo quidem, Trophime, répons – (2014), ‘San Bartolomeo d’Azzano (Asti): une tradition liturgico-musicale alto-italique atypique (890- Post domini nostris salvatoris V. Perfuderat eius, Saturnin, répons 1220)’, Rivista internazionale di Musica sacra 35/1-2, pp. 21-63. Post gloriosum salvatoris V. Qui ad sacre fidei, Trophime, répons Vidit Ihesus publicanum V. At ille relictis, Matthieu, répons – (2016), ‘Les offices ses saints et la formation du paysage liturgique provençal: modélisation et réseaux Vox Christi discipulum, Barthélémy, translation, antienne ecclésiastiques (ixe-xiie s.)’, Provence historique 65/259, pp. 39-56.

Hanssens Johann Michael ed. (1948-1950), Amalarii episcopi opera liturgica omnia, 3 vols., Città del Vaticano: Biblioteca Apostolica Vaticana (Studi e Testi 138-140).

Bibliography Hartmann-Virnich Andreas ed. (1999a), Le portail de Saint-Trophime d’Arles. Naissance et renaissance d’un chef d’œuvre de l’art roman, Arles: Actes Sud. Andrault-Schmitt Claude ed. (2006), Saint-Martial de Limoges: ambition politique et production culturelle, xe-xiiie siècles: actes du colloque tenu à Poitiers et Limoges du 26 au 28 mai 2005, Limoges: Presses – (1999b), Sur les traces des sculpteurs et maçons. Les découvertes archéologiques in Hartman-Virnich Universitaires de Limoges. 1999a, pp. 107-157.

Baudot Jules – Chaussin Léon eds. (1935-1959), Vies des saints et des bienheureux […], 13 vols., Paris: – (2015), ‘Les cathédrales provençales à l’époque de l’essor roman’, Provence Historique 65/257, Letouzey-Ané. pp. 93‑113.

Bozóky Edina (2006), La politique des reliques de Constantin à Saint Louis, Paris: Beauchesne. Jacobsson Ritva (2008), The Antiphoner of Compiègne: Paris, BnF lat. 17436 in Rebecca A. Baltzer - Margot E. Fassler eds., The Divine Office in the Latin Middle Ages: Methodology and Source Studies, Regional Bibliotheca sanctorum (1961-), Roma: Istituto Giovanni xxiii della Pontificia Università lateranense - Città Nuova. Developments, Hagiography, Oxford: Oxford University Press, pp. 147-178.

Cattin Giulio (1990-1992), Musica e liturgia a San Marco. Testi e melodie per la liturgia delle ore dal Jacobsson Ritva - Haug Andreas, Versified Office in Grove online. XII al XVII secolo. Dal graduale tropato del Duecento ai graduali cinquecenteschi, 4 vols., Venezia: Fondazione Levi (Collezione speciale per la musica veneta. A, Monumenti). Kelly Thomas Forrest (1988), ‘Neuma Triplex’, Acta Musicologica 60, pp. 1-30.

Claire Jean (1986), La musique de l’Office de l’Avent in Jacques Fontaine - Robert Gillet - Stan M. Lapidge Michael (2017), Hilduin of Saint-Denis, The Passio S. Dionysii in Prose and Verse, Leiden: Brill Pellistrandi eds., Grégoire le Grand: Chantilly, Centre culturel Les Fontaines, 15-19 septembre 1982, (Mittellateinische Studien und Texte 51). Paris: Cnrs, pp. 649-659. Lemarié Joseph (1965), Le Bréviaire de Ripoll, Paris, B.N. lat. 742. Étude sur sa composition et ses textes Duprat Eugène (1943-1946), ‘Histoire des légendes saintes de Provence: Saint Victor à Marseille’, Mémoires inédits, Montserrat: Abadia de Montserrat (Scripta et Documenta 14). de l’Institut historique de Provence 20, pp. 66-94; 21, pp. 3-40. Leroquais Victor (1934), Les bréviaires manuscrits des Bibliothèques publiques de France, Mâcon: Protat, t. 2, Gasser Sylvain (1992), ‘Les antiennes O’, Études grégoriennes 24, pp. 53-84. n. 456, pp. 431-434 (fol. 82).

e e Gilles-Raynal Anne-Véronique (2006), Le dossier hagiographique de saint Saturnin de Toulouse in Monique Lobrichon Guy (1989), La Madeleine des Bourguignons (xi -xii siècles) in Ève Duperray ed., Marie- Goullet - Martin Heinzelmann directors, Miracles, vies et réécritures dans l’Occident médiéval, Madeleine dans la mystique, les arts et les lettres. Actes du Colloque international Avignon, 20-22 juillet Ostfildern: Thorbecke, pp. 341-405. 1988, Paris: Beauchesne, pp. 71-88.

Goudesenne Jean-François (2002a), L’Office romano-franc des saints martyrs Denis, Rustique et Eleuthère, Mathon Gérard (1967), Onorato in Bibliotheca sanctorum 9, coll. 1202-1203. composé à Saint-Denis à partir de la Passio du Pseudo-Fortunat (VIe-VIIIe s.), remanié et augmenté par l’archichancelier Hilduin vers 835 puis au Xe s., Ottawa: The Institute of Mediaeval Music (Historiae Morin Germain Dom (1909), ‘La formation des légendes provençales: faits et aperçus nouveaux’, Revue Series, Musicological Studies 65/6). bénédictine 26, pp. 24-33.

374 375 CANTUS PLANUS 2014

Océane Boudeau Ogliari Francesco (2007), La vita di San Siro: vescovo e martire patrono della città di Pavia, Pavia: Selecta.

Olivier Véronique (2010), La Vie de sainte Marthe de Tarascon: édition, traduction et analyse historique, Le sanctoral de la cathédrale de Sens Dissertation, Montréal: Université du Québec (http://www.archipel.uqam.ca/2649/1/M11273.pdf).

Saxer Victor (1949a), Marie-Madeleine (22 juillet) in Baudot and Chaussin, 7, pp. 526-542.

– (1949b), Légendes provençales in Baudot and Chaussin, 7, pp. 542-543.

– (1955), ‘L’Origine des reliques de Sainte-Marie-Madeleine à Vézelay dans la tradition hagiographique du Moyen Âge’, Revue des sciences religieuses 29/1, pp. 1-18. Depuis le quatrième siècle, Sens s’est imposé comme le siège de l’archevêché de la province ecclésiastique éponyme qui comprenait notamment le diocèse de Paris. – (1958), ‘Les saintes Marie Madeleine et Marthe de Béthanie dans la tradition liturgique et homilétique orientale’, Revue des sciences religieuses 32/1, pp. 1-37. L’archevêque de Sens se trouvait donc à la tête d’un vaste territoire qui se confondait en partie avec le domaine royal. Le sanctoral de la cathédrale de Sens comprend six fêtes – (1959), Le culte de Marie-Madeleine en Occident des origines à la fin du moyen ãge, 2 vols., Auxerre: Publications de la Société des fouilles archéologiques et des monuments historiques de l’Yonne - Paris: caractéristiques, qui ne se retrouvent cependant pas toutes exclusivement à Sens. Leur Clavreuil (Cahiers d’archéologie et d’histoire 3). présence simultanée à ces dates dans un calendrier ou un sanctoral trahit cependant un manuscrit sénonais: Saxer Victor - Celletti Maria Chiara (1967), Maria Maddalena in Bibliotheca sanctorum 8, coll. 1078-1107.

Vitale Stefania (forthcoming), The Office of St Eusebius of Vercelli from the eleventh to the twentieth century: le 7 janvier: l’Invention des saintes Reliques (Inventio sanctarum reliquarum) prolegomena to an edition in David Hiley ed., Historiae: Liturgical Chant for Offices of the Saints in the Middle Ages, Venezia: Fondazione Levi. le 27 janvier: Sainte Paule (Paula)

Zeiller Jacques (1926), ‘Les origines chrétiennes en Gaule’, Revue d’histoire de l’église de France 12/54, le 28 juillet: Sainte Colombe (Columba) pp. 16-33. le 11 août: la Translation de la Couronne d’épines (Translatio sancte Corone) le 1er septembre: Saint Loup (Lupus) le 19 octobre: Saints Savinien et Potentien (Savinianus et Potentianus)

La Translation de la Couronne d’épines correspond au passage à Sens de la relique christique en provenance de Venise. Acquise par le roi Louis ix, qui chercha par la suite à se procurer d’autres reliques de la Passion pour lesquelles il fit construire l’im- mense reliquaire qu’est la Sainte Chapelle, la sainte Couronne arriva en août 1239 à Paris après avoir fait étape à Sens. Les calendriers parisiens et sénonais gardent une trace de cette cérémonie fixée au 11 août avec la fête de la Translation de la Couronne, à Sens, ou sa Réception (Susceptio), à Paris. La célébration apparaît dans les calen- driers notée par la main principale si le calendrier a été élaboré après 1239, ou ajoutée postérieurement si le manuscrit est antérieur.

Cette étude sur le sanctoral s’appuie sur un corpus de dix calendriers dont les plus anciens ont été notés vers 1200:

US-CAh lat. 429: le calendrier et les quelques folios de chants de l’ordinaire de la messe qui suivent ont été notés pour un autel de la cathédrale de Sens, probablement celui de la chapelle Saint-Michel-en-la-Tour, fondée au début du treizième siècle au

Cet article a bénéficié des conseils et remarques de Christian Meyer. Qu’il en soit remercié.

376 377 CANTUS PLANUS 2014 BOUDEAU — LE SANCTORAL DE LA CATHÉDRALE DE SENS

premier étage de la tour nord de la cathédrale (Chartraire 1921, p. 17 et 20).1 la fête à neuf leçons, celle à trois leçons avec Te Deum, trois leçons et enfin la seule «memoria». Les calendriers de US-CAh lat. 429 et F-SEm 6 font apparaître une distinc- F-AS 1005 (718): le calendrier, lacunaire, a été relié avec les fragments d’un tion plus fine entre les fêtes doubles à cinq cierges et les fêtes doubles à sept cierges. antiphonaire originaire du diocèse de Sens noté vers 1200. On voit également apparaître dans le missel imprimé de 1575, le nouveau degré de F-PR 11 (4): le calendrier précède un missel-séquentiaire à l’usage d’un autel de «semiannuale» qui se place entre les fêtes doubles et les fêtes annuelles. Plusieurs la cathédrale, peut-être l’autel de Notre-Dame. Parmi les obits qui y ont été ajoutés, fêtes, jusqu’alors doubles, deviennent désormais semi-annuelles. figurent ceux de l’archevêque Guy de Noyers, mort en 1193, et du pape Innocentiii , mort en 1216. En revanche, l’obit de l’archevêque Pierre de Corbeil, mort en 1222 n’a Les saints de la cathédrale de Sens pas été ajouté, ce qui laisserait supposer que les obits n’étaient plus actualisés à cette Après avoir soustrait le fonds commun des saints inscrits dans les calendriers gré- date, et que le manuscrit aurait été noté vers 1200. gorien et gélasien et dont le culte était universel, j’ai classé les saints présents dans F-SEm 15: le calendrier a été relié avec un missel qui a été en usage à l’autel de la les calendriers de Sens en reprenant la distinction de Victor Leroquais (1932-1934, 1, paroisse Sainte-Croix, placé à la base de la tour sud de la cathédrale. Il a été noté p. lxvi): avant 1239. les saints universels dont le culte est largement répandu: quatre-vingt-cinq saints F-Pn lat. 10502: le calendrier précède un autre missel-séquentiaire. Il a également les saints régionaux dont le culte est avéré dans le nord de la France: soixante-seize été noté avant 1239. saints F-Pn lat. 1028: le calendrier est relié avec un bréviaire. La fête de saint Guillaume de et enfin les saints locaux dont le culte est propre à Sens: cinquante saints. Bourges (canonisé en 1218) est de première main mais la fête de la Couronne d’épines a été ajoutée, ce qui laisse supposer que le calendrier a été noté entre 1218 et 1239. Mais il est également possible de classer cet ensemble de saints en s’attachant cette F-SEm 6: le calendrier débute le livre du préchantre. Il a été noté après 1239. fois-ci à leurs origines: les saints contemporains du Christ ou des apôtres; ceux ayant vécu aux premiers siècles de la Chrétienté, souvent originaires d’Orient ou d’Italie; F-SE Cerep (Centre de Recherche et d’Étude du Patrimoine): le calendrier enfin les saints originaires de Gaule. aL plupart des saints de ce dernier groupe sont précède un bréviaire. Il a été noté après 1239 et avant 1297 puisque la fête de saint originaires du nord de la France, tout particulièrement de Sens (vingt-trois saints) Louis a été ajoutée. avec les célèbres saints Savinien et Potentien (premiers évêques de Sens), saint Loup F-MOf H 71: le calendrier est relié avec un missel noté vers 1400. Le calendrier (évêque) ou encore sainte Colombe (vierge et martyre). Les autres régions ayant semble cependant plus ancien puisque pour le 25 août, la Translation de saint fourni un contingent important de saints sont – sans surprise –, les régions voisines Savinien a été notée puis barrée et la fête de saint Louis est notée mais sans le degré appartenant à la province ecclésiastique. Huit saints, dont le célèbre Germain, sont de la fête. Le calendrier aurait donc été copié peu après la canonisation du saint ainsi originaires d’Auxerre, sept de la région de Troyes et six de la région d’Orléans. (1297), au moment de l’introduction de cette nouvelle fête à Sens. Le nombre très réduit de saints nouvellement canonisés nous surprend davantage. F-Psg imprimé 4 BB 205 INV 428: le calendrier et le missel qui suit ont été imprimés La plupart des saints célébrés, y compris les saints locaux, ont ainsi vécu avant le en 1575. huitième siècle. On trouve cependant quelques fêtes nouvelles: Saint Thomas de Cantorbéry (canonisé en 1173 et fêté le 29 décembre), la Translation de la Couronne Dès le début du treizième siècle, le calendrier sénonais est stabilisé puisqu’entre les d’épines que nous avons déjà mentionnée, Saint Louis (canonisé en 1297 et fêté le 25 calendriers les plus anciens et celui de 1575, les changements sont minimes et ne août),2 Saint Roch (mort en 1380? et fêté le 16 août). Les fêtes de la Visitation (2 juillet), concernent le plus souvent que l’ajout de nouvelles célébrations ou des modifications de la Transfiguration (6 août) et même de saint François (canonisé en 1228 et fêté dans la hiérarchie des fêtes. À Sens, la fête la plus importante est la fête annuelle, le 4 août) figurent dans l’imprimé de 1575 et ont été ajoutées dans les autres livres. suivie de la fête double, puis de la fête à neuf leçons avec Te Deum. Viennent ensuite Celles de saint Guillaume de Bourges (canonisé en 1218 et fêté le 11 janvier) et de saint Pierre, «ermite et évêque» que l’on peut identifier avec le pape Célestin v canonisé 1. Calendrier décrit par Barbara Haggh-Huglo Description and Evaluation of Cambridge, MA, Harvard University, en 1313 et fêté le 19 mai, apparaissent dans quelques calendriers mais ne semblent Houghton Library, MS Lat. 4291: Complete Calendar of the Use of the Abbey of St. Pierre-le-Vif in Sens and Incomplete Kyriale (http://www.musmed.fr/AdMMAe/Haggh-Huglo,%20Calendar%20of%20Sens. pdf). Ce calendrier comprend toutes les caractéristiques des autres calendriers de la cathédrale. Cette observation est d’ailleurs confirmée par les informations contenues dans les obits. 2. À Sens, cette célébration a remplacé la Translation de saint Savinien déplacée au 23 août.

378 379 CANTUS PLANUS 2014 BOUDEAU — LE SANCTORAL DE LA CATHÉDRALE DE SENS pas avoir réussi à s’imposer à Sens.3 Dans son étude sur le sanctoral de l’abbaye méridionaux (Arles, Gèrone, Burgos) et septentrionaux (Paris, Soissons ou Chartres) Saint-Victor de Marseille, Victor Saxer (1966, p. 490-491) avait également remarqué ou de saint Caprais (Caprasius, mémoire le 20 octobre), martyrisé dans sa ville natale, un nombre important de «saints antérieurs au septième siècle», tout en précisant que Agen, en 303 (Bénédictins de Ramsgate 1991, p. 107), tous deux n’étant pas liés à la cette particularité ne signifiait en aucun cas que le calendrier s’était constitué à cette ville de Sens. En revanche, saint Gérault (Geraldus), seigneur d’Aurillac et fondateur époque, mais que la découverte et le culte des reliques avaient joyeusement brouillé de l’abbaye de cette même ville, a vu sa vie relatée par Odon de Cluny, ce qui, très les pistes, ce qui semble vraisemblablement avoir été le cas à Sens comme nous le certainement, explique sa popularité au-delà des contrées méridionales. Le séjour que verrons plus loin. fit le corps de saint Privat, évêque de Mende, à l’abbaye de Saint-Denis avant d’être rapporté en Gévaudan explique également sa popularité dans les villes du nord. Le nombre de saints relevant du monde monastique surprend également. Mais leur culte était, le plus souvent, soit largement répandu au sein de la chrétienté, soit dif- La présence des reliques d’un saint explique de façon certaine le développement d’un fusé dans d’autres localités voisines de Sens, ce qui ne rend leur présence dans les culte. Ainsi, si saint Magloire (Maglorius), évêque de Dol en Bretagne, est fêté à Sens calendriers séculiers sénonais en rien remarquable, d’autant plus que ces fêtes ne le 24 octobre (mémoire), c’est très certainement à cause de la présence de ses reliques consistent le plus souvent qu’en de simples mémoires. Ainsi, saint Maur (15 janvier), à Paris. Il en va de même de saint Philibert (mémoire le 20 août), abbé de Jumièges sainte Scholastique, sœur de saint Benoît (10 février), saint Liphard (3 juin), fonda- puis de Noirmoutier, dont les reliques ont été déplacées en Bourgogne pendant les teur-abbé de l’abbaye de Meung-sur-Loire près d’Orléans (Bénédictins de Ramsgate invasions normandes, tout comme celles de saint Siviard cité plus haut. Son culte est 1991, p. 314), saint Gilles (Aegidius, 1er septembre), abbé d’un monastère bénédictin à d’ailleurs bien diffusé au nord de la France. La présence de reliques apparaît ainsi l’endroit de l’actuelle ville de Saint-Gilles (ibidem, p. 226), saint Ayoul (Aigulfus), abbé comme le déclencheur d’un culte qui n’aurait pas forcément existé en l’absence de ces de Lérins près de Cannes (3 septembre), saint Seine (Sequanus, 19 septembre), ab- précieux restes. Cette constatation n’est bien sûr en rien propre à Sens et s’étend à tous bé-fondateur d’un monastère dans le diocèse de Langres (ibidem, p. 451), saint Bercaire les établissements religieux de la chrétienté. (Bercharius, 16 octobre), abbé-fondateur de Montier-en-Der et de Pellemoutier (ibi- dem, p. 87), saint Hilarion (21 octobre), saint Colomban de Luxeuil (21 novembre) ne Des saints, des reliques et des offices bénéficient que d’une mémoire et saint Laumer (Launomarus, 19 janvier), fondateur La fin du douzième siècle est une période de profonds remaniements à Sens puisque du monastère de Corbion près de Chartres (Body 1907, 1, p. 382-384), d’une fête à trois la construction de la façade de la nouvelle cathédrale, une des premières cathédrales leçons. Mais le culte de certains saints monastiques – peu nombreux en comparaison gothiques, s’achève. Son autel avait déjà été consacré en 1164 par le pape Alexandre iii, des précédents – n’est avéré qu’à Sens sans qu’il soit toujours possible de l’expliquer. alors en exil. À cette date, le chœur et la nef étaient probablement déjà terminés C’est le cas de saint Martin de Saujon en Saintonge, «viator» et disciple de saint Martin mais il a fallu attendre la fin du douzième siècle pour que l’édifice soit totalement de Tours (mémoire le 8 mai),4 mais aussi de saint Siviard (fête à neuf leçons, 1er mars), achevé (Chartraire 1921, p. 15). La nouvelle cathédrale constitue alors un réceptacle abbé de Saint-Calais près du Mans (Bénédictins de Ramsgate 1991, p. 465) dont la nouveau pour la liturgie quotidienne du chapitre, ainsi qu’un nouvel écrin pour les cathédrale possédait les reliques depuis les invasions normandes (Chartraire 1897, reliques. Comme ces dernières ont été déplacées pendant le chantier, elles ont fait p. 13) et de saint Paterne (mémoire le 13 novembre), moine d’origine bretonne qui l’objet d’une cérémonie de retour, d’une translation, le 16 août 1192 sous la direction avait fait partie de la communauté monastique de Saint-Pierre-le-Vif (Bénédictins de de l’archevêque Guy de Noyers.5 Le procès-verbal établi à cette occasion comptabilise Ramsgate 1991, p. 390). 133 reliques auxquelles il faut ajouter celles que possédaient les autres établissements religieux de Sens, en tout premier lieu les abbayes Sainte-Colombe qui a accueilli Ces dernières constatations peuvent s’étendre à d’autres saints d’origine non monas- saint Thomas de Cantorbéry pendant son exil et Saint-Pierre-le-Vif où vécut Odoranne tique. Ainsi, si la plupart des saints originaires de Gaule ont vécu à Sens ou de façon de Sens. La quatrième croisade, qui débute en 1202, et tout particulièrement la prise plus générale en Bourgogne ou bien dans des villes relativement proches, plusieurs de Constantinople par les Croisés en 1204, permit par ailleurs d’alimenter en objets saints sont issus de régions beaucoup plus éloignées. C’est le cas de sainte Eulalie de le Trésor. Cet enrichissement concerne notamment les tissus orientaux qui servaient Barcelone (mémoire le 12 février) qui se retrouve (ibidem) à la fois dans les calendriers alors à enfermer les reliques posées dans le reliquaire. Le second suaire de saint Potentien fait ainsi très probablement partie de ces objets rapportés par les Croisés.6 3. La fête de saint Guillaume figure ainsi dans les calendriers de F-Pn lat. 1028 et F-SE Cerep notée par les mains principales. Elle a été ensuite barrée dans F-SE Cerep et n’apparaît pas dans les calendriers postérieurs. La Saint Pierre a été ajoutée seulement dans deux manuscrits: F-SE Cerep et F-SEm 6. 5. Voir l’édition du Procès verbal de la translation des reliques de la cathédrale de Sens, opérée le 16 août 1192 par Gui de Noyers, archevêque de Sens (Chartraire et Prou 1898). 4. Voir la base de données Calendoscope élaborée par Denis Muzerelle: http://calendoscope.irht.cnrs.fr/accueil (consultée le 27/02/2015). 6. Description de la chasuble par Marielle Martiniani-Reber (Musée du Louvre 1992, p. 380).

380 381 CANTUS PLANUS 2014 BOUDEAU — LE SANCTORAL DE LA CATHÉDRALE DE SENS

La présence de ces reliques permet de comprendre l’importance de certaines célé- Trois fêtes se distinguent de toutes celles que je viens d’évoquer puisqu’elles marquent brations. En effet, ce ne sont pas seulement les reliques des saints locaux qui ont fait l’arrivée d’une relique particulière dans le Trésor. Il s’agit de la Réception d’une côte l’objet d’acquisitions, mais aussi celles de saints dont le culte dépassait les limites de de Marie-Madeleine, «Susceptio coste beate Marie Magdalene», fêtée le 14 novembre la ville de Sens. Parmi les saints locaux, les différents établissements religieux de Sens avec une fête à neuf leçons; de la Réception du pied droit de saint Étienne, «Susceptio possédaient bien sûr les reliques de plusieurs évêques, à commencer par les deux pre- peditis dextri beatissimi Stephani», ajoutée au 4 août; et bien sûr de la Translation miers évêques de Sens, saints Savinien et Potentien, dont les restes étaient conservés de la Couronne d’épines. Il est très probable que le coffret en ivoire sculpté connu à Saint-Pierre-le-Vif (Picard 1992, p. 28) et qui bénéficiaient, à la cathédrale, d’une fête sous le nom de la «Sainte Châsse», réalisé à Venise ou en Sicile au douzième ou trei- double avec sept cierges. Ces reliques côtoyaient celles d’évêques moins illustres: saint zième siècle, ait servi de reliquaire à la sainte Couronne d’épines.9 Agricius (ibidem, p. 30), trois leçons et Te Deum le 13 juin; saint Ambroise (Ambrosius, ibidem, p. 28 et 30), mémoire ou trois leçons avec Te Deum le 3 septembre; saint Léon Afin de rendre certaines de ces fêtes plus grandioses, le clergé sénonais composa plu- (Leo, ibidem, p. 28 et 30; Chartraire 1911, p. 34), trois leçons et Te Deum le 22 avril; sieurs offices. Outre l’office monastique pour les saints Savinien et Potentien contenu à saint Ursicin (trois leçons et Te Deum le 24 juillet);7 saint Héracle (fête double avec cinq la suite du traité d’Odoranne de Sens dans le manuscrit de la Bibliothèque apostolique cierges le 9 juillet) et saint Paul (fête double avec cinq cierges le 5 juillet) pour lesquels du Vatican (V-CVbav Regin. lat. 577) et très probablement noté à Saint-Pierre-le-Vif, on il ne restait qu’un suaire au treizième siècle (Chartraire 1897, p. 28); saint Arthème trouve également des offices séculierspour Sainte Paule, Saint Loup et la Translation de (trois leçons le 28 avril) dont les reliques étaient conservées à Saint-Pierre-le-Vif (Body la sainte Couronne. Ces offices comportent de nombreux répons chargés de mélismes 1907, 4, p. 685); saint Ebbon (neuf leçons le 27 août), évêque de Sens après avoir été afin de rehausser la solennité de la fête. Cette technique était très prisée à Sens, comme abbé de Saint-Pierre-le-Vif qui conservait sa chasuble (Chartraire 1897, p. 45-46; et dans bien d’autres établissements religieux à cette époque. Plusieurs des répons sont Musée du Louvre 1992, p. 378); et enfin saint Loup (fête annuelle ou double avec sept par ailleurs des contrafacta qui reprennent la mélodie de répons plus anciens. Ainsi à cierges, 1er septembre), enterré dans l’abbaye Sainte-Colombe (Chartraire 1911, p. 7) et Sens, pour l’office de Sainte Paule, les six premiers répons des matines sont empruntés dont le culte dépassait largement le cadre de la ville de Sens puisqu’il était également aux offices (F-Pn nal. 1535, fols. 126-129): fêté à Paris, Senlis, Orléans ou encore Soissons.8 À ces différents évêques s’ajoutent également quelques martyrs sénonais dont les restes étaient gardés à Saint-Pierre-le- de la Trinité: R. Sancta Paula Graecorum V. Potens quondam divitiis (contrafactum Vif (Picard 1992, p. 28 et 32): Sanctien (Sanctianus), Augustin et Béate (mémoire ou du R. Benedicat nos Deus, Cao 6240: GBEMV HRDFSL), trois leçons avec Te Deum le 6 septembre), ainsi que saint Sérotin, compagnon des de la Trinité ou (à la cathédrale de Sens) de la Toussaint: R. Ingressa Christi sepul- saints Savinien et Potentien (mémoire le 22 septembre). chrum V. Quid ibi lacrimarum (contrafactum du R. O beata trinitas, Cao 7254: DFS), de Saint Nicolas: R. Inter doloris aculeos V. Quis dabit michi pennas (contrafactum du D’autres saints font l’objet d’un culte particulier grâce à la présence de leur relique et R. Operibus sanctis, Cao 7324: BE RDFSL), R. Curiose Christi pauperes V. Exspoliabat sans que leur existence ait été particulièrement liée à la ville de Sens. C’est le cas de filios et maiorem (contrafactum du R. Qui cum audissent, Cao 7474: BE RDFSL) et saint Colloquille (Colloquillus), roi au sujet duquel on ne possède pas d’informations R. Nesciebat se matrem V. Aversos tenebat oculos (contrafactum du R. Beatus Nicholaus, mais dont le corps était conservé à la cathédrale et qui était fêté le16 mars avec une Cao 6222: BE RDFS), fête à neuf leçons; mais aussi de sainte Paule (fête double le 27 janvier), veuve romaine disciple de saint Jérôme, dont la cathédrale possédait la relique du corps entier, ou de Saint Vincent: R. Sicut inter multas gemmas V. Quanto se deiciebat tanto (mélodie encore de saint Victor (fête double avec sept cierges le 20 juillet), martyr en Thébaïde proche du R. Christi miles preciosus, Cao 6277: E DF). dont la cathédrale possédait la relique de sa tête cédée à Pierre de Corbeil en 1206 suite au sac de Constantinople (Riant 1875, p. 190-191). En revanche, saint Thomas Becket L’antienne de Magnificat des premières vêpres, O felix ancilla Dei, était également de Cantorbéry (fête double avec cinq cierges le 29 décembre), assassiné en 1170 et chantée à la cathédrale de Paris pour la fête de leur sainte locale, sainte Geneviève. canonisé trois ans plus tard, entretint avec Sens des liens particuliers puisqu’il avait Une autre fête, cette fois-ci très répandue puisqu’il s’agit de l’Assomption, a également trouvé refuge à Sainte-Colombe pendant son exil. La cathédrale de Sens conservait par fait l’objet de compositions propres. Contrairement aux trois offices cités plus haut, ailleurs un ornement liturgique lui ayant appartenu (Chartraire 1897, p. 42). l’office de l’Assomption ne comprend pas uniquement des compositions nouvelles.

7. Les reliques des saints Léon et Ursicin étaient conservées à la basilique Saint-Léon ou à Saint-Pierre-le-Vif. 9. La Sainte Châsse est décrite par Jannic Durand (Musée du Louvre 1992, pp. 264-265). Outre les dimensions de la Châsse (diamètre de 32,5 cm) qui correspondent à celles de la Couronne d’épines (28 cm), le fait que les scènes représentées 8. Voir Calendoscope (note 4). comportent «un symbolisme princier» tend à conforter cette hypothèse. Voir Musée du Louvre (2001, p. 59 et passim).

382 383 CANTUS PLANUS 2014 BOUDEAU — LE SANCTORAL DE LA CATHÉDRALE DE SENS

De plus, les chants nouveaux ont été réservés à la vigile de la fête elle-même, le 14 Anima mea liquefacta est août, la musique du 15 août constituant sûrement un noyau plus ancien. Au sein des Ora ergo pro devota matines de la vigile, les efforts des compositeurs se sont concentrés sur les répons, les Ave regina celorum (Cao 1542: D) antiennes étant des compositions largement répandues: Speciali devotione Magnificemus Deum Premier nocturne Ibi fruitur visione R. Sanctas primicias V. Non calor hunc coxit Ave magnifica misericordie R. Regni celestis V. Restitui numerum Felix commertium ubi R. Virginitas celum V. Celicus ordo Gloriose semper virginis Prosule Benedic nobis (HB 79)10 Salve regina misericordie (@Cantus: très diffusée) Prosule Benedicat omnis (HB 78) Speciosa facta es (@Cantus: GB-Cu Mm.ii.9, Augustins de St. Giles Abbey à Barnwell, Prosule Benedictus sit hodie (HB 80) deuxième quart du treizième siècle et I-AO 6, collégiale de Sant’Orso à Aoste, début du treizième siècle) Deuxième nocturne R. Porta Syon V. Intacte matri Sancta Dei genitrix virgo semper (Cao 4699: B HFSL) R. Unam quam V. Divinum (@Cantus: D-MZb D; Mayence, ca. 1430)11 Sancta Maria virgo perpetua R. Quindenis gradibus V. Post genitum In prole mater in partu (Cao 3274: CBEV HRDFL) Sub tua protectione Troisième nocturne Sub tuum presidium R. Dum deitas nati V. Ut raperet celum Sancta Maria succurre miseris (Cao 4703: BEM HDFSL) R. Ecclesie sponsum virgo V. Sic secum matrem R. Virgula thurifera V. Pulvere fumigero L’évolution du calendrier et l’adoption de nouveaux offices L’ajout d’offices dans les manuscrits permet de prendre connaissance de l’évolution du Michel Huglo avait remarqué que le sixième répons, Quindenis gradibus, se rencon- sanctoral. Dans l’antiphonaire F-Pn nal. 1535, l’office de sainte Paule n’est pas intégré trait dans une version polyphonique dans I-Fl Pluteus 29.1 (Huglo 1991, p. 156-157). parmi les autres fêtes du sanctoral. Bien qu’il soit de la main du copiste principal, il figure à la fin du manuscrit, entre le commun des saints et l’office des défunts. Cette L’antiphonaire de Sens, F-Pn nal. 1535, possède à la suite de l’antienne de Magnificat particularité laisse supposer que l’office était encore récent au moment de la rédaction des secondes vêpres, des antiennes regroupées sous la rubrique «Alie antiphone» (fols. du manuscrit. Il se peut qu’il ait été composé pour la Translation des reliques de 1192 96-98v) parmi lesquelles plusieurs sont très peu diffusées: parmi lesquelles figurait la relique de la sainte (Chartraire et Prou, p. 139). Le bréviaire F-Pn lat. 1028 donne également des informations quant à l’évolution du sanctoral. On Beata dei genitrix et semper virgo trouve ainsi après l’office des défunts qui clôturait initialement le livre, les offices de Benedicta et laudabilis virgo la Translation de la Couronne d’épines, de sainte Catherine, de sainte Anne, de la Regina mundi Maria Toussaint, de la Fête-Dieu et de saint Louis: Imperatrix celorum nos a malis libera Te laudant angeli patris fol. 286: De Translatione sancte Corone (main B, Ah 5 no. 9)12 Tota pulchra es (Cao 5162: B RS) fol. 292v: Katerine (main B, voir Ah 26 no. 69 et Ah 18 no. 38) Benedicta incomprehensibilis Dei gratia fol. 298: Anne (main C, voir Ah 25 no. 18) Alma redemptoris mater (Cao 1356: D) fol. 305: Omnium Sanctorum (main D, chants très diffusés empruntés à divers Credimus enim quia (Cao 1418: B S) offices) fol. 313: Corporis et sanguinis Domini (main E, office en prose très diffusé) fol. 328: De sancto Ludovico (main F, Ah 13 no. 71) 10. «HB» correspond au catalogue des prosules établi par Hofmann-Brandt (1971, vol. 2). Les trois prosules notées ici ne se retrouvent que dans la liturgie sénonaise.

11. http://cantusdatabase.org/ (consulté le 28 février 2015). 12. Au sujet de cet office, voir Arnaud et Dennery (2012). «Ah» correspond à Dreves, Blume et Bannister (1886-1922).

384 385 CANTUS PLANUS 2014 BOUDEAU — LE SANCTORAL DE LA CATHÉDRALE DE SENS

La main B, à qui on doit les offices de la Translation de la Couronne et de sainte Tableau 1. L’évolution du sanctoral dans les calendriers sénonais Catherine, est différente de celle qui a noté l’ensemble du manuscrit (main A) bien qu’elle en soit très proche et semble seulement légèrement postérieure à cette main principale. On peut donc en déduire que l’office de sainte Catherine a été introduit à US-CAh F-AS F-SEm F-PR F-Pn lat. F-Pn lat. F-SE F-SEm 6 F-MOf F-Psg lat. 429 1005 15 11 1028 10502 Cerep H71 imprimé la cathédrale de Sens plus ou moins en même temps que l’office de la Translation de 4 BB 205 INV 428 la Couronne, c’est-à-dire vers 1239. Cette supposition est confirmée par l’examen des vers vers avant vers 1218- avant 1239- après peu après 1575 calendriers. Si on compare les fêtes apparues à partir du treizième siècle pour nos dix 1200 1200 1239 1200 1239 1239 1297 1239 1297 calendriers, on peut les classer chronologiquement et établir ainsi l’ordre d’introduc- Janvier tion de ces fêtes à Sens (voir tableau 1). S’il est difficile de savoir quel calendrier est le 11 Guillelmi × × plus ancien parmi US-CAh lat. 429, F-AS 1005 (718), F-SEm 15 ou F-PR 11 (4), la pré- sence de Saint Guillaume (Guillelmus) de Bourges, canonisé en 1218, dans F-Pn lat. Avril 1028 permet de placer la rédaction de ce calendrier entre 1218 et 1239. Le calendrier 2 Marie Egyptiace y × × × × de F-Pn lat. 10502, dans lequel apparaît la nouvelle fête de sainte Catherine, serait pos- 30 Eutropii y y y y × × × térieur à la rédaction de F-Pn lat. 1028, mais toujours antérieur à l’introduction de la Translation de la Couronne d’épines qui n’apparaît écrite par la main principale qu’à Juillet partir du calendrier du F-SE Cerep, en même temps que Sainte Marie l’Égyptienne. 26 Etherei × lacune × × × × × × × La fête de saint Eutrope, que l’on rencontre notée par le rédacteur principal de F-SEm Lupi (de Troyes) × × × × 6, aurait été introduite à la suite de la fête de la Translation de la Couronne d’épines, donc après 1239. Anne × × Marcelli La situation semble plus complexe pour Sainte Anne dont l’introduction pourrait être 27 Hylarionis × lacune × × × × × × × × considérée comme contemporaine de celle de Saint Eutrope, si elle n’apparaissait pas pour le 26 juillet dans les calendriers de F-Pn lat. 1028 et de F-SE Cerep. Cette indéter- Anne [y ?] × × × mination laisse d’ailleurs entrevoir le processus d’introduction de cette nouvelle fête. Août

Bien que l’on rencontre cette fête habituellement le 26 juillet, ce n’est pas la date qui a 11 Translatio Corone y lacune y y × × × × été finalement retenue à Sens. On peut y voir l’influence possible du calendrier de la Tyburcii × × × × × × × × × cathédrale de Paris qui, fêtant ce jour-là la Translation de saint Marcel, a dû reporter la fête au 28 juillet, le 27 étant consacré à la fête de la Transfiguration. La cathédrale 25 Translatio × lacune × × × × [×] [×] × Saviniani de Sens ne fêtant pas la Transfiguration, elle a pu placer la fête deSainte Anne le Ludovici y y y y y × 27 juillet, d’autant plus que le 28 juillet était consacré à sainte Colombe et le 26 à saint Loup de Troyes avec une fête à trois leçons. Le 27 juillet laissait ainsi davantage de Novembre possibilités, saint Hilarion ne bénéficiant que d’une simple mémoire. 25 Petri × × × × × × y × × ×

Katerine y y y y × × × × × Cette courte présentation du sanctoral de la cathédrale de Sens a ainsi permis de mettre en avant l’évolution du sanctoral à Sens, tout en affinant la datation de certains Décembre calendriers. On notera le décalage pouvant exister, au sein d’un même livre, entre 8 Conceptio Marie × × × y × × × y × × la date de copie du calendrier et celle du bréviaire ou du missel qu’il contient. C’est ainsi le cas pour le calendrier de F-SEm 6 qui comprend la fête de la Translation de la ×: fête notée par la main principale Couronne d’épines alors que le sanctoral du corps du manuscrit qui suit ne comporte y: fête ajoutée ultérieurement pas l’office. ×: fête barrée [×] et [y]: fête grattée

386 387 CANTUS PLANUS 2014

Tova Leigh-Choate Bibliography

Arnaud Brigitte - Dennery Annie (2012), L’Office de la Couronne d’épines à Sens, Lions Bay: Institute of The weekly commemorative office: Mediaeval music (Historiae 19). a twelfth-century witness from Saint-Denis in context Bénédictins de Ramsgate (1991), Dix mille saints. Dictionnaire hagiographique, [Turnhout]: Brepols - [Québec]: Sigier.

Body Auguste ([1907-]), Vies des saints, Lyon: L’Œuvre des bonnes lectures [nouvelle édition].

Chartraire Eugène (1897), Inventaire du trésor de l’église primatiale et métropolitaine de Sens, Sens: In the fall of 1124, two years into his abbacy, Abbot Suger issued a liturgical ordinance Duchemin. for the monastic community of the royal abbey of Saint-Denis. Its primary purpose was – (1911), Les Tissus anciens du trésor de la cathédrale de Sens, Paris: Champion. twofold: to renew and augment the abbey’s weekly Saturday memorial for the Virgin and to inaugurate on Thursdays a parallel service in honor of Saint Denis and his companions – ([1921]), La Cathédrale de Sens, Paris: Laurens (Petites monographies des grands édifices de la France). Rusticus and Eleutherius. In Suger’s words (2008, 2, p. 161), this new memorial for the abbey’s «most glorious and sweetest patron» was to be «in order and equality the same» Chartraire Eugène - Prou Maurice ([1898]), ‘Authentiques des reliques conservées dans le Trésor de la (eodem ordine et eadem paritate) as that for the holy Mother of God. Suger hoped that by Cathédrale de Sens’, Mémoires de la Société nationale des antiquaires de France 59, pp. 129-172. «remembering [the saint’s] great benefices, we might have recourse to a favorable and Dreves Guido Maria - Blume Clemens - Bannister Henry Marriott eds. (1886-1922), Analecta hymnica propitious supporter on the last and terrible Day of Judgment, before the stern Judge; medii aevi, 55 vols., Leipzig: Reisland. and that we […] might successfully obtain some end of eternal happiness at the feet of Hofmann-Brandt Helma (1971), Die Tropen zu den Responsorien des Officiums, 2 vols., Erlangen: Hogl. our Lord and Master». His was a reciprocal arrangement: by offering Saint Denis the gift of a weekly liturgy in his honor, to be celebrated perpetually, the abbot sought the saint’s Huglo Michel (1991), Observations sur les origines de l’École de Notre-Dame in Christian Meyer ed., Aspects de la musique liturgique au Moyen Âge, Paris: Créaphis, pp. 151-158. intercession at the final judgment, with the ultimate hope of «finding mercy in some extremity of Paradise» (ibidem, p. 165).1 In his ordinances, Abbot Suger used the term Leroquais Victor (1932-1934), Les Bréviaires manuscrits des bibliothèques publiques de France, 6 vols., Paris: n.pr. memoria for his weekly liturgies for Mary and Saint Denis. Later manuscripts from the abbey favor the term commemoratio. Both terms, of course, had multiple meanings in Musée du Louvre ed. (1992), Byzance. L’art byzantin dans les collections publiques françaises, Musée du the medieval liturgy, and they are often, together with the even more confusing ‘votive’, Louvre, 3 novembre 1992-1er février 1993, Paris: Réunion des musées nationaux. used imprecisely in modern scholarship. What, then, did Suger intend for his weekly – (2001), Le Trésor de la Sainte-Chapelle. Paris, musée du Louvre, 31 mai 2001-27 août 2001, Paris: memorials and how did they relate to liturgical traditions at Saint-Denis and elsewhere? Réunion des musées nationaux. Chant scholars often focus on music for the prominent feasts of the liturgical calendar, Picard Jean-Charles (1992), Province ecclésiastique de Sens in Nancy Gautier - Jean-Charles Picard eds., Topographie chrétienne des cités de la Gaule des origines au milieu du VIIIe siècle, 8, Paris: de Boccard. whether of the temporale or sanctorale; these, after all, gave shape to the annual round, dictating the times of feast and fast and linking each religious community to the Riant Paul (1875), ‘Les dépouilles religieuses enlevées à Constantinople au XIIIe siècle par les latins’, Mémoires de la société nationale des Antiquaires de France 4/6, pp. 1-214. people and places of salvation history. But the ferial liturgies, the liturgies of so-called ordinary time, and the votive and commemorative forms of liturgy that were added Saxer Victor (1966), ‘Les calendriers liturgiques de Saint-Victor et le sanctoral médiéval de l’abbaye’, increasingly to the daily liturgical round – in some cases, replacing it – did as much Provence historique 16/65, pp. 463-519. or more to shape religious experience as did the annual celebration of Easter or the Assumption. Yearly and seasonal cycles played out as weeks and, within these, days of organized worship, and these daily and weekly cycles formed the warp and woof of the liturgy. What role did music play in the everyday life of religious communities? More specifically, how did Suger’s ordinance translate into actual liturgy and song, and what was its impact on the liturgical round and the liturgical community?

1. Suger’s ordinance is edited and translated in Suger (2008, 2, pp. 156-167). All English translations are my own. On Suger and reciprocity, see Maines (1986, pp. 76-94). Research for the revision of my paper was supported by a fellowship from the National Endowment for the Humanities. All conclusions are my own.

388 389 CANTUS PLANUS 2014 LEIGH-CHOATE — THE WEEKLY COMMEMORATIVE OFFICE

This essay explores the context and contents of Suger’s ordinance of 1124, its Interpreting the phrase «secundum quod in catalogo huius institutionis intitulatum relationship to weekly commemorative traditions more broadly, and (briefly) its actual est» as referring to an earlier ordinal, Anne Walters Robertson (1991, pp. 243-244) implementation at Saint-Denis, as shown in the abbey’s surviving liturgical books. It hypothesized that a commemorative office existed previously and that the purpose offers some evidence that Suger’s ordinance of 1124 may be the earliest reference to a of Suger’s ordinance was to reestablish it, perhaps after a period of disuse. By its weekly commemorative office for a patron saint other than Mary. By commemorative placement and content, however, the phrase seems instead to refer to the memorial office, I refer to a total liturgy of the hours – not just a mass – that would replace going forward, as established by Suger’s charter and inscribed in some kind of the normal ferial cursus with proper or otherwise festive material, sometimes (as at catalog of commemorations.3 Suger’s use of the three verbs continuare, attolere, Saint-Denis) even blending with festal liturgies for a weekly memorial throughout the and decorare also suggests, beyond rhetoric, some amount of augmentation and liturgical year. Abbot Suger, famous for his work as a patron of art and architecture, elaboration of a previous tradition. Edward Foley (1990, p. 129) tactfully described regent of France, and historian, emerges as a liturgist of significance, whose Suger’s «reestablishment» as both continuing and enhancing «an obscure older commitment to liturgical innovation and cultic renewal preceded the works for which tradition» of Saturday Marian worship at the abbey. Perhaps a closer look at Suger’s he is so well known: the rebuilding and refurnishing of the abbey church in what ordinance and abbey manuscripts may shed some light on this «obscure older would become known as a new Gothic style, and the texts describing these efforts.2 tradition» and give us a better understanding of the nature of Suger’s liturgical innovations. I. Marian Saturdays at Saint-Denis Suger opens his ordinance of 1124 (2008, 2, p. 159) by acknowledging the hand of The Saturday mass all-powerful God in his accession to the abbacy of Saint-Denis and in the safety with Special Marian commemoration at Saint-Denis dates back at least as far as Abbot which he has been able to serve the Church. His indebtedness, he writes, prompts him Hilduin, who around 832 built a chapel honoring Mary and all the saints in the crypt to exalt the Lord and to support the monastic community in its exercise of the divine of the Carolingian basilica and called for the continuous performance there of the service. His first matter of business is a confirmation and re-envisioning of the abbey’s day and night office. Robertson (1991, p. 224) noted that ninth-century portions of Saturday memorial to the Virgin: the manuscript F-R 275 (A. 566) contain a series of votive masses for Mary and all the saints possibly used in this chapel. Among several votive mass formularies for Unde, ad honorem Dei omnipotentis et beatae Dei genitricis semperque virginis Mariae, in capitulo Mary found in a ninth-century sacramentary made for Saint-Denis F-Pn lat. 2290 nostro generaliter residentes, ipsius Sanctae Dei genitricis memoriam continuare, attollere, are one for Saturdays (fol. 135) and one that matches the commemorative mass as decorare constituimus, eo videlicet tenore ut deinceps aeternaliter, secundum quod in catalogo 4 hujus institutionis intitulatum est, omni die sabbati solenpniter celebretur quemadmodum in outlined in the abbey’s surviving manuscripts (fol. 126). This latter formulary (collect octavis Pentecosten tribus extremis diebus, preter quod septem psalmos cum letania et vigiliis Concede nos famulos tuos, secret Tua Domine propitione, and postcommunion Sumptis mortuorum dimitti prohibemus. Domine salutis) is also highlighted twice, alongside additional prayers and lessons, in the eleventh-century votive material in the Rouen manuscript mentioned above (fols. Whence, for the honor of the omnipotent God and the blessed and ever virgin Mary, mother of God, we, sitting in our general chapter, have determined to continue, to augment, and to ornament 12v-13 and 28v-29), unfortunately both times without rubric. The same formulary is the memorial of this same holy mother of God, by this observance, namely, that the [memorial] be singled out as the only Marian mass in the votive material of the eleventh-century celebrated solemnly every Saturday, in a continuous series forever, according to what is inscribed sacramentary-gradual, F-Pn lat. 9436 (fol. 129); although not specifically identified in the catalog of this institution, in the same manner as in the last three days of the octave of in relation to a Saturday tradition, the mass receives extra prominence through an Pentecost, except that we prohibit the omission of the seven psalms, together with the litany and accompanying half-page illumination of a standing Virgin flanked by two angels. vigils of the dead. Contemporary manuscripts from other liturgical centers, moreover, do associate this From this, and from later references in Suger’s ordinance and other charters, it is clear Marian mass formulary with a Saturday commemorative tradition (see, for example, that Saturday was to be a «day of holy commemoration» (sanctae memoriae dies) – a the Bury missal from ca. 1120, F-LA 238, fol. 149). feast, of sorts. Suger’s mention of the Pentecost octave and, in particular, the penitential psalms, litany, and office of the Dead, confirms that his vision for the memorial 3. For a similar use of the term catalogus in reference to the record of a commemorative liturgy established by charter, see included a full liturgy of the hours. Bijsterveld (2007, p. 164). In a survey of the term catalogus in medieval texts, I found no examples referring to an ordinal or like manuscript; rather, the term consistently applies to a list or group of notices or texts such as virtues or vices, bishops, saints, nations, or, as in this case, commemorations. 2. For more on Suger’s liturgical involvement and further bibliography, see Leigh Choate (2009) and Robertson (1991). On Suger’s legacy in his various spheres, see Gerson (1986), Große (2004), Bur (1991), and Grant (1998). I am currently 4. Among these are the earliest extant ordinal, F-Pm 526, fol. 83 (which gives three possible prayer formularies), and the completing a book that focuses more specifically on liturgy and the saints at Suger’s Saint-Denis. unnotated missal, F-Pn lat. 846, fol. 59-61, both from the thirteenth century.

390 391 CANTUS PLANUS 2014 LEIGH-CHOATE — THE WEEKLY COMMEMORATIVE OFFICE

More indicative at Saint-Denis is the presence of a special mass proper in commemora­ Pentecost week suggest that he raised the memorial’s rank enough to keep it regularly tione sanctae Mariae in the abbey’s earlier eleventh-century gradual, F-Pm 384 in the round, in contrast with many centers, where coincidence with even three-lesson (fols. 136v-137), not within a separate votive section but within the temporale itself, feasts might cause its omission or displacement. following masses for the post-Pentecost Sundays and the Holy Trinity. The chants of this mass, beginning with the introit Salve sancta parens, closely match those in the The Saturday office Saturday mass outlined for the post-Pentecost ordinary time in the abbey’s thirteenth- Less certain than the pre-existence of a Saturday mass is the question of whether, century ordinal (F-Pm 526, fol. 83; Foley 1990, p. 440), and reflect a wide Saturday before Suger, the abbey’s weekly commemoration included a communal liturgy of Salve tradition, evident in numerous manuscripts including the Bury missal noted the hours in addition to the mass, or whether Suger was the first to add the Saturday above.5 Characteristic of such weekly commemorative liturgies, the Saint-Denis Marian office. Earlier manuscripts show no evidence of a commemorative office. gradual even gives two of the four alleluias listed in the later ordinal. Thus, by the Even a handful of Marian hymns, tantalizingly placed just after those for Pentecost eleventh century a votive mass for Mary seems to have been introduced formally into in the psalter-hymnal F-Pn lat. 103 (fols. 154v-155), seem to have been intended for the abbey’s round, as a communal commemoration in choir, presumably – though it the annual Marian feasts and do not (with one exception) correlate with those found is not stated – on Saturdays. Robertson (1991, p. 300) surmised that the Saturday mass later in the commemorative office at Saint-Denis. In his charter Suger allocated an at Saint-Denis would have been celebrated most likely in Hilduin’s chapel in the crypt increased meal allowance for Saturday, likewise suggesting a significant change in the or in the Marian chapel in Suger’s new chevet. As the main mass of the day, however, day’s liturgy commensurate with the introduction of a festal office. it was likely sung in choir, and the later ordinal gives no indications to the contrary. The daily Marian commemorations, on the other hand, were generally performed in By Suger’s time, it seems, relatively few communities in the Latin west celebrated Suger’s chevet and, before that, may have been performed in the crypt. a weekly commemorative office for Mary. First attested in the late tenth-century customary of Einsiedeln, which mandated three-lesson Friday and Saturday offices By the 1120s, a weekly Marian mass on Saturdays, while not universal, was not for the Cross and the Virgin, respectively, the custom was beginning to take root by uncommon. The Regularis Concordia of ca. 970, likely based on continental practice, the second half of the eleventh century (Clayton 1990, p. 88). Texts for the Marian ordered Anglo-Saxon monasteries to celebrate the main mass for Mary on Saturdays office (and that for the Cross) are first extant in the so-called Portiforium of Saint and for the Holy Cross on Fridays, unless coinciding with a feast (Symons 1953, p. 20). Wulstan (GB-Ccc 391), a composite office book from Worcester compiled ca. 1065 Cluniac liturgy, as represented by the Liber tramitis (ca. 1040), specified only a Friday (Hughes, 2, 1960; Clayton 1990, pp.77-78; Roper 1993, pp. 94-95, 242). Mary Clayton Mass for the Holy Cross (Dinter 1980, p. 121), but Bernard tells us, in his customary has argued for the origin of these three Marian lessons in Winchester in the first half written around 1080, that Abbot Hugh (1049-1109) introduced a main mass for Mary of the eleventh century, before their spread to Worcester and other places in England on Saturdays during ordinary time, or, if Saturday was occupied by a feast, on another and on the continent by mid-century (Clayton 1990, p. 88). A late eleventh-century open day (Herrgott 1726, p. 347). Not surprisingly, the early twelfth-century ordinal legendary from Saint-Sépulchre in Cambrai contains both the undivided lessons and from Rheinau (CH-Zz Rh. 80), which witnesses the Hirsau reform liturgy inspired by a text added slightly later, Quare omni sabbato memoria sanctae Mariae celebratur, Cluny, similarly prescribed weekly commemorative masses for Mary and the Cross which (as its name suggests) argues for the solemn celebration of a Saturday Marian for the periods between the octave of Pentecost and Advent and the octave of Epiphany office, in addition to the festive mass with Gloria already found in many churches.6 The and Septuagesima; when coinciding with feasts, however, these would be omitted appearance of this propagandistic text in wide-ranging manuscripts from the eleventh (Hänggi 1957, p. 173; Heinzer 1992). Many twelfth-century manuscripts, including and, especially, twelfth centuries, including from Saint-Maur-des-Fossés and Saint- the Bury missal noted above and an ordinal from Montecassino (V-CVbav Urb. lat. Martin-des-Champs, demonstrates growing interest in the weekly office and, more 585), witness a cycle of commemorative ferial masses for the long weeks of ordinary broadly, the cause of Marian celebration across a wide geographical area, including time following the Pentecost octave, including masses for the Holy Cross on Friday the Paris region. and Mary on Saturday (Kelly 2008, p. 415); the Bury missal gives full texts for both. Abbot Suger’s ordinance for Saint-Denis likely augmented a Saturday tradition similar Abbot Suger may have known the lessons and Quare omni sabbato, not to mention the to these. Together with the witness of later manuscripts, Suger’s mandate that the prayers and other Marian devotions spreading rapidly on both sides of the channel. memorial be celebrated continuously and his comparison with the final three days of His ordinance recalls several themes common to contemporary mariology: that one

6. Barré (1967, p. 379) suggested a continental origin, noting that Udon, bishop of Toul from 1052-1069, championed a 5. For more evidence of this tradition, see Frénaud (1961, pp. 190-191), and Ihnat (2014, pp. 74-75). Saturday Marian office, and possibly Gerard of Csanad (d. 1046) as well.

392 393 CANTUS PLANUS 2014 LEIGH-CHOATE — THE WEEKLY COMMEMORATIVE OFFICE serves God by serving his sacrosanct mother Mary, that one’s efforts to honor the and seek the mater misericordiae, as encouraged in the Quare omni sabbato and other queen of angels and men remain insufficient, yet worthy of God, and that through the contemporary Marian texts. continuous suffrage of the holy mother of God one can merit forgiveness of sins and, ultimately, a place in Paradise. By the thirteenth century, and possibly in Sugar’s time It could be, as Robertson suggested, that a Saturday office already existed at the abbey, as well, the Saturday office at Saint-Denis used the same three lessons found in the perhaps even as far back as the 1060s (around the time of the Worcester Portiforium); Portiforium (Sacrosanctam venerabilis Dei genetricis).7 Suger’s comparison of his new- Suger’s ordinance, in this case, would have served to confirm the memorial, elevate and-improved memorial to the final three days of the Pentecost octave suggests such its festivity, endow it in perpetuity, and attach to it his own hopes for salvation in a three-lesson format in his day as well. The eleventh-century gradual-antiphoner (F- the form of extra Saturday psalms on his behalf. However, Suger’s own language, Pm 384) and later ordinals show that the ferias within Pentecost week, like Pentecost the manuscript evidence, and the wider context of Marian devotion at the abbey and and its Easter model, were celebrated festively at the abbey in the Roman manner, beyond suggest that it was he who transformed a memorial mass into an especially with three lessons at matins.8 Such a comparison may also have served to raise the festive three-lesson Saturday ‘feast’ for the Virgin Mary. This, if true, would suggest a memorial’s festivity in other ways. Abbey ordinals prescribe for the final three days of level of continuity in the commemorative office tradition not heretofore recognized. the Pentecost octave the ringing of all the bells at vespers and lauds (sonet classicum), Roper noted that the ‘pre-conquest’ office found in the Worcester Portiforium seems the singing of certain chants by one or two soloists, censing by a priest in copes, the use to have fallen out of favor in England in the immediate aftermath of the Norman of the Sunday psalms, and other marks of festivity (Foley 1990, p. 425).9 conquest; she identifies the next two witnesses of the Marian office in English manuscripts – from Saint Alban’s and Winchcombe, both from around the time of Suger’s mandate to retain the daily chanting of the seven penitential psalms and Abbot Suger – as having a twelve-lesson format which would only yield again to litany and the vigils of the dead, at a time when (presumably) they would have been three lessons in the thirteenth century (Roper 1993, pp. 96-103). The Gesta Abbatum omitted, suggests that the character of the commemoration was to be both festive and of Saint Alban’s credits the Norman abbot Geoffrey Gorron, Suger’s contemporary, penitential. Indeed, to aid his quest for mercy in the day of Resurrection, Suger ordered with introducing the twelve-lesson Saturday office there (ibidem, p. 97). Some of the the singing of an extra penitential psalm on his behalf after every liturgical hour on manuscripts cited by Barré also reveal a full festal format of nine or twelve lessons both Saturdays (Mary’s day) and Thursdays (Saint Denis’s day), for the rest of his life (Barré 1967, p. 376). Suger’s clear prescription of a three-lesson format suggests that and following his death. the two traditions may be more parallel than previously thought; could their format possibly be related to the parallel traditions of three-and twelve-lesson Easter and […] omni quinta et septima feria qua prefatae celebrari poterunt memoriae, ad peccatorum Pentecost feasts (Hesbert 1982)? meorum depositionem miserrimaeque vitae hujus directionem, in omnibus tam nocturnis quam diurnis horis, psalmum unum, Ad te levavi, et post miseri corporis hujus dissolutionem, in ipsis sanctarum memoriarum diebus, De profundis clamavi, per omnium curricula temporum II. Suger’s Thursday Feast for Saint Denis misericorditer fidissima et irrefragabili promissione mihi meisque successoribus obtinuimus If we must hesitate regarding Suger’s contributions in the Marian realm, we can be (Suger 2008, 2, p. 165). fairly certain that it was he who introduced at the abbey the commemorative office for Saint Denis on Thursdays and possibly the weekly mass as well. His phrase «[…] hujus The abbot clearly wanted his own memory tied to that of his two primary patrons; sanctissimae Dei genitricis memoriam eodem ordine et eadem paritate in quinta feria in this way, Suger’s memorial psalms and commemorations prefigured the visual secundamus» is somewhat ambiguous, however, for the rare verb secundare can mean and epigraphic images he later had placed around his new abbey church (Maines «to favor or further» but also «to adapt or adjust favorably». In any case, growing devotion 1988). The most famous of these is the Annunciation panel in the Infancy window to the Virgin Mary clearly inspired Suger’s cult-building projects for Saint Denis, and of the axial chapel in Suger’s new chevet, in which the abbot is shown prostrate considering that even the Saturday Marian office needed justification, Suger’s parallel and barefoot at the feet of the annunciate Mary. Here, with his Ave Maria, the Angel Thursday office – of the same order and equality (eodem ordine et eadem paritate) – Gabriel teaches the abbot (and, by extension, the monastery and monarchy) to honor seems nothing if not radical. In his brief sketch, Foley (1990, p. 120) calls the Thursday commemoration «unquestionably, one of the most distinctive features of the liturgy at 7 The earliest ordinal, F-Pm 526 (fol. 82v), and the Saturday office found in the liturgical fragments following the Saint-Denis». Perhaps the significance extends beyond the royal abbey, however. To my illuminated Life of Saint Denis in F-Pn n.a.fr. 1098 (fols. 60v-61v). knowledge, no earlier examples of weekly commemorative offices for patron saints 8. On this tradition of celebrating a Roman-cursus matins for Easter and Pentecost, see Foley (1990, pp. 146-150) and Hesbert (1982). (beyond Mary) have been identified, and the few scholars who have written about such

9. For the degrees of festivity at the abbey related to censing, see Foley (1995, pp. 193-195). offices have dated their appearance to the thirteenth century (Tolhurst 1993, p. 150;

394 395 CANTUS PLANUS 2014 LEIGH-CHOATE — THE WEEKLY COMMEMORATIVE OFFICE

Foley 1990, p. 121; and, especially, Roper 1993, pp. 113, 239-241).10 Thursday mass in the abbey’s ordinal F-Pm 526 (fol. 82) and incomplete missal F-Pn lat. 846 (fols. 61-63), both from the thirteenth century. By Suger’s time, too, the brothers Surely many factors inspired Suger, in the autumn of 1124, to raise the liturgical of Saint-Denis likely chanted daily commemorative antiphons at the end of vespers profile of Saint Denis at the abbey. These may have included the contest in 1121 over and lauds. This practice was promoted in England by the Regularis Concordia (ca. the saint’s Areopagite identity by the then-monk of Saint-Denis, Peter Abelard, which 970) and can be seen at Christ Church, Canterbury and Cluny in the eleventh-century apparently resulted in his temporary exile from the kingdom (Leigh Choate 2009, pp. (Tolhurst 1993, pp. 101-103); later ordinals from Saint-Denis outline the practice 124-157); Suger’s extensive travels in 1123 to the important cult centers of Rome and probably instituted there before Suger’s time. Even the abbey’s unique anniversary southern Italy, where he may have witnessed weekly ferial mass cycles highlighting office for King Dagobert, which Suger likely helped establish before his abbacy, may 11 saints other than Mary; and the presumed role of Saint Denis in the recent bloodless have played a part (Barroux 1942-1943). victory of Louis vi’s forces over Emperor Henry v, who had threatened to invade from Germany. In his biography of Louis vi, Suger (1964, p. 228) notes that the relics of Saint Suger’s role thus seems to have been to transform a supplementary votive mass for Denis and companions, displayed on the altar throughout this campaign, were visited Saint Denis, prayed privately or communally on irregular occasions (and possibly by «crowds of devout people and pious women» who came to add their own prayers to Thursdays), into an official commemorative mass sungin conventu as the main or, the continuous office performed there «day and night» by the brothers. Exaggeration less frequently, morning, mass on an almost weekly basis, and to augment the short, or not, it is reasonable to assume that the monks of Saint-Denis would have compiled supplementary daily commemorations for Saint Denis with a full weekly liturgy of and celebrated extra liturgies in the martyrs’ honor during this time and that the the hours. By likening the Thursday memorial to that for the Virgin on Saturday, offices’ apparent efficacy, and the intensity of popular devotion toward the saint, may «celebrated solemnly as in the last three days of the Pentecost octave», Suger (2008, have inspired Suger’s liturgical undertakings. Certainly, too, the king’s increasing 2, p.160-61) gave the Saint Denis memorial a festive three-lesson format rivaling reliance upon Saint Denis as his patron and protector – a relationship dramatically the abbey’s twelve-lesson feasts. He also augmented the monks’ food allowance on enhanced through Suger’s own efforts – justified regular liturgical service in the Thursdays and Saturdays in perpetuity, «whether they perform [the memorials fully], saint’s honor. Resemblances between Suger’s ordinance and two conciliatory royal or suitably adapt (mutare) them on the occasion of the vigils of some high-ranking diplomas attributed not just to Suger’s influence but possibly to his own hand show solemnity or Lent».13 Suger’s memorials were not simply additions to the day’s liturgy; that the relationship between monarchy and monastery, between royal ceremonial Thursdays and Saturdays thenceforth assumed the role of memorial days, not only and monastic liturgy, was very much on the abbot’s mind in the first half of the 1120s.12 in commemoration of Saint Denis and the Virgin Mary but also, through the addition of memorial psalms in his own honor, the abbot himself. Later manuscripts show Abbey traditions themselves, some centuries old, likely influenced both the day and the that the memorial liturgies consisted of variable, composite offices, with chants drawn content of Suger’s weekly commemoration. Saint Denis and companions are invoked at from the annual offices of Mary and Saint Denis, and arranged with some flexibility in (or near) the head of a series of special patrons named in a missa specialium sanctorum the choice of certain elements. found among the votive masses assigned to Thursdays in the ninth- and tenth-century sacramentaries F-Pn lat. 2290 (fols. 129v-130), and F-LA 118 (fol. 35). Although the Conclusion eleventh-century gradual F-Pm 384 contains no formulary for an ordinary-time mass Weekly commemorative liturgies seem to have developed as part of the efforts of for Saint Denis (as it did for Mary), the later eleventh-century sacramentary F-Pn lat. religious communities to order and give meaning to the ferial liturgies of the ‘dull 9436 does include within its vast collection of votive masses a formulary specifically seasons’ of ordinary time.14 Their introduction as communal ritual reflected prevailing honoring Saints Denis, Rusticus, and Eleutherius (fol. 131v), which (although not tied currents in theological culture, popular devotion, and even politics. In some centers, to Thursdays or highlighted in any way) matches the first formulary listed for the weekly commemorations involved only the mass; in others, the office as well. A weekly office reflected a higher degree of festivity than a weekly mass alone; at Saint-Denis,

10. Parkes (2014, p. 151) suggests that a short office headed in uigilia sancti Eadmundi in the early twelfth-century Bury libellus, US-NYpm M.736, may be a single-nocturn weekly commemorative office of the type described by Sally Roper, though he concedes that it was more likely intended for use on the day before Edmund’s primary feast, as suggested 13. Gasparri translates Suger’s verb mutare as déplacer, suggesting that the abbot intended the memorials to be moved to by its rubric. Lapidge (1991, p. lxix) mentions the attribution to Aethelwold of Winchester of a private, supplementary another day. However, abbey practice documented in the later ordinals shows that the memorials were only rarely omitted office for Saints Peter and Paul, listed in a text called De horis peculiaribus copied into the eleventh-century manuscript completely and were never transferred to another day, both common practice in other institutions; instead, the weekly F-AL 14; such a private office would be more of a votive or ‘little office’ than a commemorative communal office. memorials were simply adapted, reduced perhaps to a morning mass and commemoration after vespers, or to the third nocturn of an important saint’s office. 11. See, for instance, the ferial mass cycles outlined in the twelfth-century ordinals of Montecassino and Benevento (Kelly 2008, pp. 324-325, 415). 14. Pfaff (1999) discusses the development of ‘sample weeks’ for the ‘dull seasons’ of ordinary time in some early English office manuscripts. The weekly commemorative mass cycle offered a related solution for the ferial masses of ordinary time, 12. On Suger’s influence in the two royal charters, see Dufour (2004, pp. 15-16). and weekly commemorative offices gave the ferial office liturgy similar structure and significance.

396 397 CANTUS PLANUS 2014 LEIGH-CHOATE — THE WEEKLY COMMEMORATIVE OFFICE the thirteenth-century ordinals show weekly commemorative masses on five of the Bibliography six ferias, but parallel offices – those for Saint Denis and Mary – on only two. In some centers, such weekly liturgies remained confined to the periods between the octave of Barré Henri (1967), ‘Un plaidoyer monastique pour le samedi marial’, Revue Bénédictine 77, pp. 375-99. Epiphany and Septuagesima and between the octave of Pentecost and Advent, while at Barroux Robert (1942-1943), ‘L’Anniversaire de la mort de Dagobert à Saint-Denis au XIIe siècle, charte others, as at Saint-Denis, they permeated the annual round more thoroughly, whether inédité de l’abbé Adam’, Bulletin philologique et historique, pp. 131-151. (in Suger’s words) celebrated fully or adapted for the occasion. Limits on time and Bur Michel (1991), Suger, abbé de Saint Denis, régent de France, Paris: Perrin. space prohibit here a full exploration of the weekly commemorative tradition at Saint- Denis and elsewhere. Some issues awaiting further exploration include relationships Bijsterveld Arnoud-Jan (2007), Do ut des: Gift Giving, «Memoria», and Conflict Management in the Medieval between weekly and annual liturgies (including new composition vs compilation), Low Countries, Hilversum: Verloren. seasonal cycles and other kinds of variability, questions of performance practice, Clayton Mary (1990), The Cult of the Virgin Mary in Anglo-Saxon England, Cambridge: Cambridge effects on and interactions with the non-commemorative ferial liturgies and other University Press. 15 votive traditions, and the related expansion of Marian and other sequence repertories. Dinter Peter ed. (1980), Liber Tramitis Aevi Odilonis Abbatis, Siegburg: Schmitt (Corpus Consuetudinum Monasticarum 10). This essay has revealed Abbot Suger as a bold and creative liturgist who understood Foley Edward B. (1990), The First Ordinary of the Royal Abbey of St.-Denis in France (Paris, Bibliothèque the fundamental power of liturgical music to shape identity and maintain memory in Mazarine 526), Fribourg: The University Press (Spicilegium Friburgense 32). religious communities. Indeed, preoccupied as he was not only with his own memory – (1995), ‘The Treasury of Saint-Denis according to the Inventory of 1234 (Paris, Bibliothèque Mazarine and eternal fate but with the intertwined legacies of the monastery and monarchy, 526)’, Revue Bénédictine 105, pp. 167-199. Suger enhanced or established numerous commemorative traditions at the royal abbey and, in some cases, its dependencies. Beyond the weekly memorials, these included Frénaud Georges (1961), Le culte de Notre Dame dans l’ancienne liturgie latine in Hubert Du Manoir ed., Maria. Études sur la Sainte Vierge, Paris: Beauchesne, pp. 157-211. the annual anniversaries of the abbey’s Merovingian ‘founder’ King Dagobert and King Louis vi, monthly commemorations for the abbey’s Carolingian benefactor Charles the Dufour Jean (2004), Suger, personnage complexe in Rolf Grosse ed., Suger en question: Regards croisés sur Saint-Denis, München: Oldenbourg, pp. 11-20. Bald, and the various liturgies he endowed for his own memoria (often connected in some way to these other commemorations). Like Guillaume Machaut and his brother Gerson Paula ed. (1986), Abbot Suger and Saint-Denis: A Symposium, New York: Metropolitan Museum of Art. Jean, who over a century later apparently attached to Guillaume’s Saturday Messe Gransden Antonia ed. (1973), The customary of the Benedictine Abbey of Bury St Edmunds in Suffolk (from de Nostre Dame a commemorative oration for themselves (Robertson 2002, pp. 269- Harleian MS.1005 in the British Museum), London: Bradshaw. 270), Abbot Suger, through his weekly commemorations and added memorial psalms, Grant Lindy (1998), Abbot Suger of St.-Denis: Church and State in Early 12th-Century France, London-New hitched his wagon to the star of the sea and mother of mercy, Mary, as well as to Saint York: Longman. Denis, his patron from oblate to abbot. It was their support that Suger anticipated on the great and terrible Day of Judgment, when he would kneel before the ‘Strict Judge’ Hänggi Anton (1957), Der Rheinauer Liber ordinarius (Zürich Rh 80, Anfang 12. Jh.), Fribourg: Éditions universitaires / Freiburg: Universitätsverlag (Spicilegium Friburgense 1). – as he later depicted on his front portal – and be given his eternal reward. Heinzer Felix (1992), ‘Der Hirsauer Liber Ordinarius’, Revue Bénédictine 102, pp. 309-347.

Herrgott Marquart ed. (1726), Bernardi Ordo Cluniacensis in Vetus disciplina monastica […] Paris: Osmont, pp. 133-364.

Hesbert René-Jean (1982), ‘Les Matines de Pâques dans la tradition monastique’, Studia Monastica 24, pp. 311-348.

Hughes Anselm ed. (1958-1960), The Portiforium of St. Wulstan, 2 vols., London: Bradshaw.

Ihnat Kati (2014), Marian Miracles and Marian Liturgies in the Benedictine Tradition of Post-Conquest England in Matthew M. Mesley - Louis E. Wilson eds., Contextualizing Miracles in the Christian West, 1100-1500: New Historical Approaches, Oxford: Society for the Study of Medieval Languages and Literature, pp. 63-98.

15. See Leigh-Choate (2009, pp. 130-153) for a more thorough discussion of the variable contents of the commemorative Kelly Thomas F. (2008), The Ordinal of Montecassino and Benevento: Breviarium sive Ordo Officiorum, 11th liturgies for Saint Denis and Mary at Saint-Denis, in relation to some of these issues. I am currently working on a wider study of sequence composition within the weekly commemorative mass tradition. Century, Fribourg: Éditions universitaires / Freiburg: Universitätsverlag (Spicilegium Friburgense 45).

398 399 CANTUS PLANUS 2014

Melanie Batoff Lapidge Michael ed. (1991), Wulfstan of Winchester: The Life of St. Aethelwold, Oxford: Clarendon.

Leigh Choate Tova A. (2009), The Liturgical Faces of Saint Denis: Music, Power, and Identity in Medieval The Visitatio sepulchri as a Gospel Harmony France, Ph. D. dissertation, Yale University. in medieval Germany Maines Clark (1986), Good Works, Social Ties, and the Hope for Salvation: Abbot Suger and Saint-Denis in Paula Gerson 1986, pp. 76-94.

Parkes Henry (2014), St. Edmund between Liturgy and Hagiography in Tom License ed., Bury St. Edmunds and the Norman Conquest, Woodbridge: Boydell, pp. 131-159. Around the year 1100, a new variety of the Visitatio sepulchri emerged in southeastern Pfaff Richard W. (1999), The Sample Week in the Medieval Latin Divine Office in Robert N. Swanson ed. Germany, one that shared little musically or textually with its precursors. This new Continuity and Change in Christian Worship. Papers read at the 1997 Summer Meeting and the 1998 Winter Meeting of the Ecclesiastical History Society Woodbridge: Boydell (Studies in Church History type of Visitatio, which flourished in Bavaria, the Austrian duchies, and later in the 35), pp. 78-88. Kingdoms of Poland, Bohemia, and Hungary, has received little attention in the

Robertson Anne Walters (1991), The Service Books of the Royal Abbey of Saint-Denis: Images of Ritual and secondary literature, despite its survival in almost 300 medieval manuscripts and Music in the Middle Ages, Oxford: Clarendon. prints and its distinctive characteristics.1 The few scholars who have discussed this new Visitatio have focused on two areas: transmission, through detailed studies of musical – (2002), Guillaume de Machaut and Reims: Context and Meaning in His Musical Works, Cambridge: Cambridge University Press. or textual variants (Norton 1983, pp. 27-28; de Boor 1976, pp. 19-20), or narrative and dramatic aspects (Young 1933, 1, pp. 239-240). Questions about how and why a new Roper Sally Elizabeth (1993), Medieval English Benedictine Liturgy: Studies in the Formation, Structure and Content of the Monastic Votive Office c. 950-1540, New York: Garland. type of Visitatio emerged, when earlier Visitationes were already known in southeast Germany, have remained unasked and will be the focuses of this paper. Symons Thomas ed. (1953), Regularis Concordia: The Monastic Agreement of the Monks and Nuns of the English Nation, London: Nelson. These newer German Visitationes must first be differentiated from earlier ones through Suger (19642), Vie de Louis VI le Gros, Henri Waquet ed., Paris: Belles Lettres (Classiques de l’histoire au nomenclature and identifying characteristics. Michael Norton’s categorization scheme Moyen Âge 46). will be adopted here; he grouped Visitationes as one of two types: type one incorporated – (20082), Œuvres, 2 vols., Françoise Gasparri ed., Paris: Belles Lettres. the famous Quem queritis in sepulchro dialogue, whereas type two featured a new version of the dialogue, Quem queritis o tremule mulieres (Norton 1987, pp. 133-137). Type one was disseminated widely both east and west of the Rhine and was first documented in the tenth century, whereas type two circulated only east of the Rhine and was first documented around the year 1100. A further difference is that type-one Visitationes featured the Marys and angels at the tomb but not the apostles, whereas type two typically included Peter and John, in addition to the Marys and angels. Mary Magdalene and the Risen Christ are additional biblical figures that occasionally appear in both Visitatio types.

The sources of the chants accompanying the dialogue and characteristics of the melodies further differentiate the two types. With few exceptions, type one and type two incorporated different collections of chants with the former one employing traditional office antiphons and the latter featuring mostly newly composed antiphons. Table 1, p. 402, lists the most frequently occurring chants in German type-one and type-two Visitationes with Cao numbers identifying office antiphons.

1. This number excludes all type-two Visitationes post-dating 1500 documented by Lipphardt (1975-1990); Carr and Norton (1993, pp. 83-90); and Evers and Janota (2013).

401 CANTUS PLANUS 2014 BATOFF — THE VISITATIO SEPULCHRI AS A GOSPEL HARMONY IN MEDIEVAL GERMANY

Table 1. Comparison of Chants in Type-one and Type-two Visitationes often did. Second, Ad monumentum venimus employs a Gallican cadence at the end of the first phrase on gementes, as the cadential pitch is approached from a step below Type 1 Type 2 rather than above. Third, there is rapid movement from the lower ambitus to the upper, Quis revolvet nobis (cf. Cao 2697) Maria Magdalena et alia Maria (no Cao) occurring at the end of the second phrase and beginning of the third with the leap of Quem queritis in sepulchro (no Cao) Quis revolvet nobis […] quem tegere (no Cao) a fifth between D and a, followed by an ascending third to c. These non-Gregorian Venite et videte locum (Cao 5352) Quem queritis o tremule mulieres (no Cao) characteristics firmly place type-two Visitationes in a different compositional stratum Cito euntes (Cao 1813) Ad monumentum venimus gementes (no Cao) from type one, further differentiating them. Surrexit Dominus (Cao 5079) Currebant duo simul (Cao 2081) Surrexit enim (Cao 5082) Cernitis o socii (no Cao) Surrexit enim (Cao 5082) Another difference between the two Visitatio types involves the sources of the chant Christ ist erstanden (no Cao) texts – a distinction that I will argue is key to understanding why type-two Visitationes were created. Type-one Visitationes recounted the Marys’ visit according to Matthew and Mark, whereas type two conflated events and quotations from all four Gospels. To illustrate this point, tables 2 and 3 present the texts of representative type-one and The office antiphons are found in some of the earliest chant sources, including the type-two Visitationes. In both examples, biblical quotations are underlined and the ninth-century Compiègne antiphoner (F-Pn, lat. 17436) and the Metz tonary (F-ME, sources are indicated in parentheses following the chant texts. 351), suggesting that they were not composed specifically for type-one Visitationes. Table 2. Type-one Ahrweiler Visitatio (Ahrweiler, Sankt Laurentius, Cod. 2a), fols. 133v-134 By contrast, many of the chants comprising the type-two Visitatio were likely composed expressly for it. The antiphons Maria Magdalena et alia Maria, Quis revolvet nobis […] R. Dum transisset sabbatum Maria Magdalena et Maria Iacobi et Salome emerunt aromata ut quem tegere, Ad monumentum venimus gementes, and Cernitis o socii had no place venientes unguerent Ihesum alleluia (Mark 16, 1) in the office and are first documented in the context of the Visitatio. The melodies of these antiphons further differentiate them from the antiphons comprising type one. V. Et valde mane una sabbatorum veniunt ad monumentum orto iam sole They might best be described as non-Gregorian, as they exhibit many stylistic traits (Mark 16, 2) that Roman Hankeln (2008, pp. 173-176) and David Hiley (1996, p. xxv; 2004, p. 371) associated with late offices, which were composed from the tenth century onward. marie dicant Ad monumentum venimus exhibits three characteristics of non-Gregorian chant (see Quis revolvet nobis lapidem ab hostio monumenti alleluia alleluia (Mark 16, 3) example 1).

Angeli Example 1. Ad monumentum venimus (A-KN, CCL 589), fol. 2, transcribed by Melanie Batoff Quem queritis in sepulchro o christicole (Matthew 28, 5; Mark 16, 6; Luke 24, 5)

marie Ihesum Nazarenum crucifixum o celicole (Mark 16, 6)

Angeli Non est hic surrexit sicut predixerat ite nunciate quia surrexit de sepulchro (Matthew 28, 6; Mark 16, 6; Luke 24, 6)

Angeli Venite et videte locum ubi positus erat Dominus alleluia alleluia (Matthew 28, 6) First, the melody frequently repeats the final and reciting tones at beginnings and ends of phrases: D and A in this case. As such, the melody remains firmly grounded on D marie throughout, rather than exploring other modal areas as traditional office antiphons Surrexit Dominus de sepulchro, qui pro nobis pependit in ligno alleluia, alleluia, alleluia

402 403 CANTUS PLANUS 2014 BATOFF — THE VISITATIO SEPULCHRI AS A GOSPEL HARMONY IN MEDIEVAL GERMANY

Table 2 is a type-one Visitatio recorded in an Ahrweiler antiphoner, Codex 2a, dated mulieres around the year 1400. The Visitatio that it transmits has the same selection and Ihesum Nazarenum [crucifixum querimus] succession of chants as much earlier Visitationes, including one found in an early (Mark 16, 6; cf. Matthew 28, 5) twelfth-century Siegburg customary (D-Mbs, clm. 14765) copied from a late eleventh- Angelus 2 century source. The sequence of events follows Mark and Matthew. As the Marys Non est hic quem queritis travel to the sepulcher with spices to anoint Christ, they ask themselves who will roll (Matthew 28, 6; Mark 16, 6; Luke 24, 6) back the stone and discover it has already been rolled back (a detail that comes from Mark). Next they encounter angels, who tell them that Christ has risen and direct them et cum ceperit cantare angelus […] to spread the news (as Matthew and Mark reported). The Marys then announce the sed cito euntes [nunciate discipulis eius et Petro quia surrexit Ihesus] (Matthew 28, 7; cf. Mark 16, 6) resurrection to the disciples (as Matthew relayed), bringing the Visitatio to a close.

Most of the chant texts are direct quotations from Matthew’s or Mark’s Gospels, namely […] Cantant mulieres Dum transisset nobis (Cao 6565), Quis revolvet nobis lapidem (cf. Cao 2697), and Venite Ad monumentum venimus gementes [angelum Domini sedentem vidimus et dicentem quia et videte locum (Cao 5352); only Surrexit Dominus (Cao 5079) is non-scriptural. The surrexit Ihesus] textual basis of the Quem queritis is more complex, as it combines short scriptural (cf. Matthew 28, 2; Mark 16, 5; John 20, 11) quotations with non-scriptural material. One may notice Luke indicated as a biblical tunc chorus imponat antiphonam source and indeed two short quotations in the Quem queritis are common to the three Currebant duo simul [et ille alius discipulis precucurrit cicius Petro et venit prior ad monumentum] synoptic Gospels, specifically the word queritis in the angels’ initial question and the (John 20, 4) angels’ proclamation «Non est hic». Since both quotations are found not only in Luke, but also in Matthew and Mark, including them does not introduce new quotations, Et […] petrus et johannes […] cantant events, or details from Luke. All the borrowings in type-one Visitationes can be Cernitis o socii ecce linteamina [et sudarium et corpus non est in sepulchro inventum] accounted for in Matthew and Mark’s accounts. (cf. John 20, 7; cf. Luke 24, 12)

chorus By contrast, type-two Visitationes incorporated events and details from all four Gospels. Surrexit enim sicut dixit [Dominus et precedet vos in Galileam alleluia ibi eum videbitis alleluia Table 3 is a type-two Visitatio from a twelfth-century ordinary from Salzburg Cathedral alleluia alleluia] (A-Su, II 6). The rubrics are abbreviated with ellipses. Broken underlining indicates (Matthew 28, 6-7) paraphrases of details or events, as does the abbreviation cf.; square brackets identify text incipits supplemented by a later Salzburg source (A-Wn, lat. 1672). The selection populus Christ ist erstanden von der marter [alle des solln wir alle froh sein Christ will unser trost sein and order of chants is typical of the other extant type-two Visitationes. Kyrie eleison]

Table 3. Type-two Salzburg Visitatio (A-Su, II 6, fol. 67)

[…] Chorus cantare inceperit The first part of the Visitatio conveys events that Matthew and Mark described: the Maria Magdalena [et alia Maria ferebant diluculo aromata Dominum querentes in monumento] Marys bring spices to the sepulcher to anoint Christ, ask who will roll back the stone, (cf. Matthew 28, 1; cf. Mark 16, 1; cf. Luke 24, 2) and encounter an angel who proclaims the resurrection and instructs them to tell the disciples. The second part of the Visitatio recounts events found only in Luke and tres presbiteri […] cantant Quis revolvet nobis [ab ostio lapidem quem tegere sanctum cernimus sepulchrum] John: on learning of the resurrection, Peter and John run to the sepulcher where they (Mark 16, 3) discover Christ’s abandoned grave clothes and handkerchief. In this manner, type-two Visitationes combine events from all the Gospels. Angelus Quem queritis o tremule [mulieres in hoc tumulo gementes] The anonymous creator or creators of type-two Visitationes also conflated the four (Matthew 28, 5; Mark 16, 6; Luke 24, 5; cf. John 20, 11) Gospels in subtler ways. Some chant texts incorporated quotations and details from more than one Gospel, which was only possible because the chant texts paraphrased 2. D-Mbs, clm. 14765, Consuetudines Sigiberti abbatis, was copied in Regensburg around 1126-1130 from an exemplar dated back to ca. 1090 (Evers and Janota 2013, 2, pp. 552-553; Lipphardt 1975-1990, 6, p. 338). rather than quoted the Gospels. Quem queritis o tremule mulieres, for example, drew on

404 405 CANTUS PLANUS 2014 BATOFF — THE VISITATIO SEPULCHRI AS A GOSPEL HARMONY IN MEDIEVAL GERMANY

Matthew, Mark, and John to differing degrees. The angel’s question to the Marys, «quem resurrection accounts and were troubled by them. After referring to the practice of queritis o tremule mulieres in hoc tumulo gementes» (whom do you seek, o trembling reading different accounts of the resurrection on consecutive days, Augustine addressed women, in this tomb, weeping?) describes the Marys weeping, a likely allusion to his congregation, stating: «[Because] you must often be admonished and must remember John’s account of Mary Magdalene crying at the tomb. The angel’s instruction for the that whatever one Gospel writer says ought not disturb you if the other Gospel writer Marys to proclaim the resurrection, given in table 4, interpolates a quotation from omits something, because he who omits that which the other says, says something Mark into a longer quotation from Matthew. that the other had omitted. […] The authority of the Holy Gospel is so great because one spirit speaks through the Gospel writers, so that whatever sort of thing even one Table 4. Texts of Matthew 28, 7 and Mark 16, 7 evangelist said is true».3 Augustine’s message was clear: the Gospels were authoritative and did not disagree; each evangelist’s account was simply incomplete. Augustine fully Matthew 28, 7 Mark 16, 7 Final line of Quem queritis o explored these ideas in an exegetical treatise, De Consensu Evangelistarum (written tremule mulieres et cito euntes dicite discipulis sed ite et dicite discipulis Sed cito euntes nunciate around the year 400), addressed to Christian readers. In it he identified and invalidated eius quia surrexit et ecce eius et Petro quia praecedit discipulis eius [et Petro] quia each seeming discrepancy, arguing that all apparent contradictions vanished when praecedit vos in Galilaeam ibi vos in Galilaeam ibi eum surrexit Ihesus. the accounts were read together. To prove his point, he incorporated short Gospel eum videbitis ecce praedixi vobis videbitis sicut dixit vobis Harmonies into his text, including a harmonized account of the Marys’ visit of the sepulcher. Augustine explained his process of harmonizing the resurrection accounts, In Matthew’s account, the angel identifies the disciples as the intended audience stating, «therefore, let us arrange all these things, which happened around the time of for the Marys’ announcement, whereas the two-word interpolation from Mark – the resurrection of the Lord, according to the testimonies of all the Evangelists, into one 4 «et Petro» – singles Peter out, so that the dialogue reads «announce to His disciples certain narrative […] as the events may have taken place». The process that Augustine and Peter». Making special reference to Peter, who figures prominently in Luke and describes is likely similar to what was involved in creating type-two Visitationes. John’s accounts, ensured continuity between the dialogue and the apostles’ scene at There was a particularly venerable tradition of harmonizing the Gospels in medieval the sepulcher and helped to draw together the four accounts. The preceding tables Germany, making it highly likely that the anonymous creator of the type-two Visitatio demonstrate that type-two Visitationes sharply contrasted with type-one Visitationes, had encountered Gospel Harmonies. The earliest evidence for their use in Germany as they conflated the four Gospels into one narrative and incorporated newly composed dates from the middle of the eighth century and is in connection with saint Boniface antiphons in non-Gregorian style. By contrast, type one recounted the Marys’ visit (672/5-754), one of the most famous missionaries to Germany and an esteemed according to Matthew and Mark and employed mostly traditional office antiphons. teacher and commentator on scripture. According to the monks of Fulda, Boniface had a copy of Tatian’s Diatessaron in his possession when he was martyred in Frisia in 754. At this point, it is important to mention that no direct evidence exists concerning why The manuscript, a mid sixth-century source of southern Italian provenance (D-FUl, type-two Visitationes may have been created. The fact that they combined the four Bonifatianus 1), apparently arrived at Fulda with Boniface’s corpse and remained Gospels into one narrative, however, may offer an important clue. The technique there to this day (Jakobi-Mirwald 1993, pp. 15-21; Raaijmakers 2012, p. 81). Given of conflating the Gospels into one account was well known in the Middle Ages. As that Boniface’s mission in Germany involved eliminating unorthodox beliefs and early as the second century, Christian theologians created Gospel Harmonies, works practices that had proliferated among recently converted Christians and clergy, one that combined Matthew, Mark, Luke, and John’s accounts of Christ’s life, death, can imagine that a Gospel Harmony could have been extremely useful for pedagogical and resurrection into one narrative. Gospel Harmonies were composed and created purposes (Raaijmakers 2012, pp. 20-23; Reuter 1980, p. 80). If anyone were to raise throughout the Middle Ages (and beyond) and were written or translated into a wide concerns about apparent conflicts among the accounts, Boniface had recourse to a range of western and eastern languages, including Latin, Greek, Old High German, compelling harmonized account. The Diatessaron was a powerful testament that the Old Saxon, Syriac, and Arabic (Petersen 1994, pp. 448-489). These works were written Gospels agreed and were authoritative. to solve a perplexing theological quandary: how could the Gospels be authoritative and true when they differed? While the earliest and most widely circulated Gospel 3. «Quod saepe admonendi estis, et memoriter tenere debetis non vos debet movere quod alius evangelista dicit, si quid Harmony, Tatian’s Diatessaron, written ca. 165-180, was an apologetic work aimed at alius praetermittit, quia et ille qui praetermittit quod alius dicit, dicit aliquid quod ille praetermiserat […] Auctoritas autem pagans and other detractors, many Gospel Harmonies were written for Christians. A tanta est evangelii sancti, ut quia in eis loquebatur Spiritus unus, verum sit etiam quod dixerit unus» (Migne 1841, col. 1117). All translations are mine. sermon delivered by saint Augustine (354-430) to his congregation, sometime after 4. «Quapropter haec omnia, quae circa tempus resurrectionis Domini facta sunt, secundum omnium evangelistarum the year 391, demonstrates that early Christians recognized the differences among the testimonia in una quadam narratione […] quemadmodum geri potuerint, ordinemus» (Weihrich 1904, pp. 361-362).

406 407 CANTUS PLANUS 2014 BATOFF — THE VISITATIO SEPULCHRI AS A GOSPEL HARMONY IN MEDIEVAL GERMANY

In ninth-century Germany, new Gospel Harmonies were invented in the vernacular, performances were sung. After all, Otfrid’s Gospel Harmony could hardly have and Tatian’s Diatessaron was copied in Latin and translated into Old High German. competed against the abominable song of the laity if it were merely read aloud. Two new vernacular Gospel Harmonies were created during this period: the anonymous poet’s Heliand (ca. 830), written in Old Saxon, and Otfrid of Weißenburg’s The presence of musical notation in the Heidelberg manuscript (D-HEu, Pal. Lat. Evangelienbuch (863-871), composed in Old High German. Two manuscripts of 52), copied in the third quarter of the ninth century in Weißenberg, offers the most Tatian’s Diatessaron (CH-SG, 56; D-Kl, 2° theol. 31) offer evidence that the text was persuasive evidence that Evangelienbuch was sung. German unheightened neumes copied in ninth-century Germany, sometimes with the Latin and Old High German are given for two rhyming couplets on fol. 17v. According to Michael Klaper (2004, translations in parallel columns, as in the case of CH-SG, 56.5 Most of these sources p. 150), the neumes likely dated to approximately the same time as the rest of the were either copied at Fulda, or written by former students of Hrabanus Maurus (789- manuscript. Many syllables are set to only one neume, but there are also compound 856), esteemed theologian and teacher at Fulda. neumes comprising as many as six pitches at the ends of hemistiches. The presence of these short neumatic passages implies that the melody accompanying Evangelienbuch The anonymous poet’s Heliand and Otfrid’s Evangelienbuch are particularly was more elaborate than a simple lectionary tone, but the unheightened notation relevant to the study of the Visitatio, as markings in the manuscripts suggest that makes it more difficult to ascertain much more. The manuscript also exhibits they were sung aloud publically. Ninth-century sources of the Heliand, written two other types of markings suggesting that Evangelienbuch was intended to be around 830, and Otfrid of Weißenberg’s Evangelienbuch, written between 863 and performed aloud: Romanian letters indicating changes in the rate of text delivery, 871, contain brief passages of unheightened German neumes and other evidence of such as t for trahere (slowing down) and c for cito (speeding up), and liquescences sung performance (D-Mbs, cgm 25 and D-HEu, Pal. Lat. 52 respectively). The focus alerting the reciter to diphthongs and consecutive liquid consonants.7 When here will be on Evangelienbuch, which was written in long-line poetry. Otfrid, a considered altogether, Otfrid’s references to «cantus» in his letter, the presence of Benedictine monk, scribe, and former student of Maurus, composed the text and neumes, and the inclusion of other performance symbols are strong indicators that oversaw its copying at the Benedictine monastery of Weißenberg (Archibald 2004, a German practice of singing Gospel Harmonies existed. pp. 140-143). A letter that Otfrid wrote to Liutbert, Archbishop of Mainz (863-889), to accompany a copy of Evangelienbuch offers compelling evidence that the work Given that type-two Visitationes and German Gospel Harmonies not only conflated was intended to be sung aloud. After explaining that he wrote in Old High German the four Gospels, but were also sung aloud publically, one must seriously consider to make the Gospels accessible to those who struggled with Latin, he described his the possibility that Gospel Harmonies were precursors to type-two Visitationes. work as if it were an antidote to the secular songs of the laity. In this letter, Otfrid Such an explanation would improve on Carl Lange (1887, p. 17) and Karl Young’s described what precipitated his writing of Evangelienbuch as follows: earlier theories (1933, 1, p. 307) that type-two Visitationes were mere expansions of type one, which is implausible given how different the two are. When on a certain occasion the sound of ineffective singing struck the ears of certain men of the highest esteem and the abominable song of the laity disturbed their sanctity – having been asked I suspect that the creator or creators of the type-two Visitatio were Augustinian by certain brothers worthy of memory, and especially by the words of a certain revered noble canons living within the Salzburg Archdiocese, given the provenances of the sources lady, by the name of Judith, who entreated me greatly – I wrote part of the Gospels in German for transmitting the type two and how strongly their transmission is associated with them in order that at least some part of this singing of the scripture may obliterate the ineptitude Archbishop Konrad of Salzburg (1075-1147) and his implementing the Rule of saint of untrained lay voices, and having been overtaken by the sweetness of the Gospels in their own language, they may learn to avoid the sound of useless singing.6 Augustine for clergy throughout the Salzburg Archdiocese. Ample evidence exists that Gospel Harmonies were copied and transmitted within the Salzburg Archdiocese Otfrid’s references to sound, voices, and ears indicate that he is contrasting two oral during this period. In fact, the creators of type-two Visitationes could have been performances, one involving the laity and another involving his Evangelienbuch. exposed to harmonization in at least three ways. First, Augustine’s De Consensu Moreover, in this context, the use of the word «cantus» suggests that the two Evangelistarum circulated widely and is listed in medieval library catalogues from the area, including two twelfth-century library lists from the city of Salzburg, as well as those from Admont, Reichenau, St. Gall, and Prüfening. Manuscripts of De 5. D-Kl, 2° theol. 31 and CH-SG, 56 were both likely copied at Fulda (Wiedemann 1994, pp. 37-38; Petersen 2004, p. 87). Consensu Evangelistarum also survive from both Salzburg Cathedral and St. Peter’s 6. «Dum rerum quondam sonus inutilium pulsaret aures quorundam probatissimorum virorum eorumque sanctitatem laicorum cantus inquietaret obscenus a quibusdam memoriae dignis fratribus rogatus, maximeque cuiusdam venerandae matronae verbis nimium flagitantis nomine Iudith, partem evangeliorum eis theotisce conscriberem ut aliquantulum huius cantus lectionis ludum saecularium vocum deleret et in evangeliorum propria lingua ocupati dulcedine sonum inutilium 7. A-Wn, 2687 has three passages marked with Romanian letters c and t: chapters I, 23; II, 3; and II, 4 (Kleiber 2004, 1, pp. rerum noverint declinare» (A-Wn, Codex 2687, fol. 4 ). 143-146). A-Wn, 2687 and D-HEu, Pal. Lat. 52 included pronunciation markings throughout (Klaper 2004, p. 148).

408 409 CANTUS PLANUS 2014 BATOFF — THE VISITATIO SEPULCHRI AS A GOSPEL HARMONY IN MEDIEVAL GERMANY

Abbey in Salzburg. Second, the creator or creators could also have encountered Bibliography harmonization through Hrabanus Maurus’ Expositio in Matthaeum or Bede’s commentaries on Mark or Luke, all of which incorporated numerous lengthy Archibald Linda (2004), Otfrid of Weissenburg in Brian Murdoch ed., German literature of the early middle quotations from Augustine’s De Consensu Evangelistarum. Third, they may have ages, Rochester, NY: Camden House, pp. 139-156. learned harmonization through Otfrid’s Evangelienbuch, as one of the sources Boor Helmut de (1967), Die Textgeschichte der lateinischen Osterfeiern, Tübingen: Niemeyer (Hermaea. (D-Mbs, cgm. 14) was copied within the Archdiocese at Freising. Germanistische Forschungen. Neue Folge 22).

Type-two Visitationes served as powerful testimony that the four Gospels made Carr Amelia - Norton Michael (1993), ‘New sources for the Visitatio sepulchri at Kloster­neuburg’, Early Drama, Art, and Music 15/2, pp. 83-90. sense when considered together and were in no way discordant. They functioned as Gospel Harmonies but offered a more immediate and potent means of asserting Evers Ute - Janota Johannes eds. (2013), Die Melodien der lateinischen Osterfeiern, 2 vols., Berlin: De Gruyter. the infallibility of scripture than earlier Gospel Harmonies. Since the events of the Marys’ visit were reenacted publically, type-two Visitationes afforded those Hankeln Roman (2008), Old and new in medieval chant: finding methods of investigating an unknown region in Ole Kongsted - Michael Kube - Niels Krabble - Morten Michelsen eds., A due: musical celebrating Easter Sunday matins the opportunity not only to hear, but also to see essays in honour of John D. Bergsagel & Heinrich W. Schwab, Kopenhagen: The Royal Library - Section the harmonized account of Christ’s resurrection unfold before their eyes. This of Musicology, University of Copenhagen, pp. 161-180. merger of two favorite German practices – harmonizing the Gospels and reenacting Hiley David ed. (1996), Historia Sancti Emmerammi Arnoldi Vohburgensis circa 1030, Ottawa: Institute of the Visitatio sepulchri – created an effective and affective way of commemorating Mediaeval Music (Musicological Studies 65/2). Christ’s resurrection, the high point of the Christian year. – (2004), Early cycles of office chants for the feast of Mary Magdalenein John Haines - Randall Rosenfeld eds., Music and medieval manuscripts: paleography and performance, Burlington: Ashgate, pp. 369‑399.

Jakobi-Mirwald Christine ed. (1993), Die illuminierten Handschriften der Hessischen Landesbibliothek­ Fulda, Stuttgart: Hiersemann.

Klaper Michael (2004), Musikhistorische Interpretationen in Kleiber 2004, vol. 1, part 2, pp. 148-153.

Kleiber Wolfgang ed. (2004), Otfrid von Weißenburg: Evangelienbuch, 2 vols., Tübingen: Niemeyer, Band 1: Edition nach dem Wienerkodex 2687, unter Mitarbeit von Rita Heuser, Teil 1: Text, Teil 2: Einleitung und Apparat.

Lange Carl (1887), Die lateinischen Osterfeiern: Untersuchungen über den Ursprung und die Entwickelung der liturgisch-dramatischen Auferstehungsfeier, München: Stahl.

Lipphardt Walther ed. (1975-1990), Lateinische Osterfeiern und Osterspiele, 9 vols., Berlin: De Gruyter.

Migne Jacques Paul, ed. (1841), Opera omnia sancti Aurelii Augustini: Sermones ad populum, Paris: Garnier (Pl 38).

Norton Michael L. (1987), ‘Of «stages» and «types» in Visitatione Sepulchri’, Comparative Drama 21, pp. 34‑61; 127-144.

Petersen William L. (1994), Tatian’s Diatessaron: its creation, dissemination, significance, and history in scholarship, Leiden: Brill.

Raaijmakers Janneke (2012), The making of the monastic community of Fulda, c. 744- c. 900, Cambridge: Cambridge University Press (Cambridge Studies in Medieval Life and Thought. Fourth Series).

Reuter Timothy (1980), The greatest Englishman: essays on St. Boniface and the church at Crediton, Exeter: Paternoster.

410 411 CANTUS PLANUS 2014

Alison Altstatt Weihrich Franz ed. (1904), Sancti Aureli Augustini opera: De consensu evangelistarum libri quattuor, Wien: Tempsky (Corpus Scriptorum Ecclesiasticorum Latinorum 43). Singing the saints in Medieval Eichstätt: Wiedemann Konrad (1994), Manuscripta theologica: die Handschriften in Folio (Die Handschriften­ der Gesamthochschul-Bibliothek Kassel - Landesbibliothek und Murhardsche Bibliothek der Stadt Kassel the case of Wolfhard of Herrieden’s office for Saint Walburga 1,1), Wiesbaden: Harrassowitz.

Young Karl (1933), The drama of the medieval church, 2 vols., Oxford: Clarendon.

The eleventh-century chronicler Anonymus Haserensis was a canon of the house of Herrieden who may have served as a cantor in the Eichstätt cathedral. In his De Gestis Episcoporum Eistetensium, the anonymous author describes Reginold, bishop of Eichstätt from 966-991, as the composer of proper offices for Saint Nicholas, for the brothers, Anglo-Saxon missionaries and Eichstätt patrons, Willibald and Wunnebald, and lastly, one for Saint Blaise, long presumed lost (Weinfurter 1987, pp. 47-48; Wendehorst 2006, p. 47). Regarding the office for Saint Willibald, the chronicler described how

at ends of certain responsories, he added very long series of notes, and added verses to these notes in the manner of the sequence […] to the ninth [responsory], he assigned the largest portion, with which he wanted to express the character of the pilgrimage to Jerusalem […] thus our exceedingly skillful musicus assembled first verses in Latin, then in Greek, from there, in Hebrew, and then again in Greek, and finally, in Latin (Jones 1963, pp. 69-70).

Dörr, Schlager, and Wohnhaas correlated this description with an office for Saint Willibald recorded in the tenth-century portion of the manuscript D-TRb Abt. 95, Nr. 5 (hereafter Trier 5) and most notably, with its extraordinary trilingual prosula.1 They furthermore proposed that Reginold wrote this office for the occasion of Willibald’s translation in April 989 to a new chapel at the western end of the cathedral (Dörr, Schlager and Wohnhaas 1987, p. 37). The two remaining proper offices in Trier 5 have attracted far less attention.2 These are for Willibald’s brother Wunnebald, first Abbot of Heidenheim, and for their sister Walburga, named abbess of Heidenheim following her brother’s death (Weinfurter 1987, p. 133; Altstatt 2014, pp. 275-279). Part of a longer study on the proper offices of Eichstätt, this paper focuses on the office for Saint Walburga. Few have noticed that Anonymus Haserensis attributed the office for Saint Walburga to the earlier poet, Wolfhard of Herrieden, who flourished at the turn of the tenth century. The chronicler relates:

This study was completed with the support of a grant from the Graduate College of the University of Northern Iowa. I thank Mother Franziska Kloos, Abbess of the Saint Walburga Abbey, Eichstätt, for generously allowing me to study the abbey’s antiphoners, and the staff of the Universitätsbibliothek Eichstätt-Ingolstadt for their assistance in accessing sources in the library’s special collections. I am grateful to the staff of the Hill Manuscript Museum and Library of St. John’s University, Minnesota, and the Bistumsarchiv, Trier, for furnishing me with photos of the manuscript D-TRb Abt. 95, no. 5. Finally, thanks are due to Dr. Hendrik Vanden Abeele for kindly sharing with me his digital images of the antiphoner NL-ZUa 6.

1. For further discussion of the prosula Terminus et idem interminus, see Hoffman-Brandt (1971).

2. A manuscript description and complete transcription of the three liturgical offices contained in D-TRb Abt. 95, no. 5 may be found at cantusdatabase.uwaterloo.ca.

412 413 CANTUS PLANUS 2014 ALTSTATT — SINGING THE SAINTS IN MEDIEVAL EICHSTÄTT

During the time of this bishop [Erkanbald] and at his order, the priest Wolfhard […] a monk of p. 176). Thus for this eighth-century witness, it was not only the vocal celebration Herrieden, edited a truly useful passionale in as much as it contained entire feasts for every single of Christ’s miracles, but also the sound of the chanted office, that marked the day of the year, and also set forth the life of Saint Walburga in four books, as instructed by the same successful establishment of Christianity. According to Huneberc, Walburga became bishop. Subsequently, after having committed a grave offence against the bishop, he was jailed, abbess of Heidenheim in the year 761, following the death of Wunnebald. She died and had no one to plead on his behalf. «I myself» he said «shall be my own intercessor». Having circa 780 and was buried in the Heidenheim abbey church (Holzbauer 1972, pp. composed historica carmina while in jail, he was at last permitted audience with the bishop. He chanted a new responsory for the holy virgin at the top of his voice, and gained not only pardon, 58-59; Weinfurter 2011, pp. 188-189). A century after Walburga’s death, Bishop but also honor and recompense (Weinfurter 1987, pp. 46-47). Otkar saw fit to translate her remains from Heidenheim to Eichstätt (ibidem, p. 197). Wolfhard’s description of Walburga’s translation resonates both with the aural While Wolfhard’s «liber passionalis» and his Miracula sanctae Walburgae have been markers of Christian liturgy. He reports that a host of dignitaries, including Liubila, identified and studied, the «historica carmina» described by Anonymus Haserensis the noble founder the nearby women’s house at Monheim, attended the exhumation, have never been correlated with manuscript evidence. Based on textual and stylistic so that «the holy remains of the virgin might be raised with the greatest of care, and characteristics, I submit that the office for Saint Walburga transmitted in Trier 5 that they might be transported to the Eichstätt cloister accompanied by the singing is most likely the same office the chronicler knew and attributed to Wolfhard. An of hymns and psalms» (Bauch 1979, pp. 158-160). Once in Eichstätt, the procession examination of its origin within the early flowering of Walburga’s cult, in which two was met with «the melodies of bells, resounding to heaven, and the sweet sound of women’s houses laid competing claim to her miraculous relics, furthermore sheds spiritual songs». As the people shed «tears of joy» Walburga’s relics were interred in light on the arrangement of its text, its musical style, and arguably, its meaning. a small chapel dedicated to the holy cross, attended by a community of canonesses The office of Saint Walburga both celebrates the saint’s recent and sensational (Bauch 1979, p. 160; Weinfurter 1987, p. 157; Weinfurter 2011, p. 197). Scarcely a posthumous miracles, and domesticates her cult by framing it against the orthodox decade after Walburga’s initial translation to Eichstätt, Liubila, founder and prioress and model of female monastic sanctity found in Wolfhard’s vita. Moreover, by of Monheim, successfully petitioned Bishop Erkanbald for a portion of her relics, retelling the miracles in alternation with the vita, Wolfhard grounds the dual promising in exchange that her foundation would become the property of the diocese aspects of Walburga’s sanctity within the sung liturgy, whose sound served as an following her death (ibidem, pp. 160-162). Wolfhard describes how in the year 893 the aural marker of Christian conversion. This liturgical retelling of Walburga’s story saint’s tomb was once again opened, accompanied by «unceasing psalms and hymns» justified her cult theologically and integrated it within a clerically-sanctioned form sung by the bishop’s entourage. As her relics began their journey to Monheim, «the of liturgical practice. Finally, the styles in which the office is written complement people of Eichstätt, some in grief, some in fear and sadness» protested their removal, its dual function of celebrating the miraculous while demonstrating an orthodox believing that «the Lady Walburga was being completely taken away from them» model of female monasticism. Seen from this perspective, Reginold’s later offices (ibidem, p. 162; Wood 2006, pp. 352-353). for Willibald and Wunebald found in Trier 5 comprise the concluding members of a trilogy: one that elevated the status of the two brothers, whose fame was previously From the very day of Walburga’s translation to Monheim, miracles began to occur. eclipsed by that of their celebrated sister, and claimed the legacy of this trio of Liubila’s house quickly became a pilgrimage site that attracted thousands seeking saints for the diocese. This study contributes to our understanding of the role of healing from the virgin Walburga. Not more than two years later, Bishop Erkanbald the proper office – not only its texts, but also its compositional styles and the sound charged Wolfhard with composing Walburga’s vita and recording her miracles (Bauch of its performance – in defining the sacred history of a locality, in re-ordering the 1979, p. 142). This task was complicated by the fact that almost nothing had been dynamics between its institutions, and in promulgating orthodoxy. recorded about Walburga’s life: thus, her vita was a blank slate on which Wolfhard fashioned the image of an orthodox female saint: a model of chastity, humility, The diocese of Eichstätt was established circa 740 in a region strategically located obedience, and religious devotion. In Wolfhard’s description of the miracles at between Bavarian, Alemannic, and Franconian lands. Its cathedral was served by Monheim, the abbey’s chapel resonates with very different sounds than the liturgical secular canons from the houses of Herrieden and later, Heidenheim (Weinfurter «hymns and psalms» that attended her translations. By contrast, he describes pilgrims 2011, pp. 190-191). In her 778 Hodoeporicon (travelogue) of Willibald’s journeys, who proclaim their suffering, beg for healing, and praise Walburga in a style that is the Anglo-Saxon nun Huneberc of Heidenheim describes Eichstätt and its environs, anything but liturgical. Wolfhard relates the lament of a certain Frankish noble woman christianized under Willibald, whence «antiphons now resound, sacred lessons are who, seeking healing for her ailing children with words full of sadness, assembled this chanted, a noble throng of believes shout aloud the miracles of Christ and with joyful mournful song, crying in a miserable voice: hearts echo from mouth to mouth triumphant praises of their creator» (Talbot 1954,

414 415 CANTUS PLANUS 2014 ALTSTATT — SINGING THE SAINTS IN MEDIEVAL EICHSTÄTT

Woe is me, that the womb of my mother brought forth a woman who should bear such pain! Woe such a way as to alternate between these three layers of time. Beginning and ending is me, that the marriage bed took away the honor of my pure innocence! Woe is me, that while I the office are the Magnificat antiphons of first and second vespers: prayers that ended my mother’s suffering with my birth, I, after becoming a mother, should bring the same take place in the here and now. The intervening office of matins alternates between pain of birth upon myself! It would be better for me, I confess, to bury many people everyday than three chronologies: the series of nine antiphons, which narrate Walburga’s vita – to have to watch, with downcast eyes, the death of a beloved child! the distance past – and the series of nine responsories, which begin and end with «In this manner», Wolfhard continues, «she arranged her words, as is the custom prayers in the present day (responsories 1.1; 3.1-3.3) which frame the recent past of among bitterly lamenting women» (Bauch 1979, pp. 214-216). The woman then her posthumous miracles (responsories 1.2-2.3). Thus the sung liturgical office, in vowed to God and to the Virgin to make a financial gift to the church if her child was which antiphons alternate with responsories, balances the sensational miracles of healed. He recovered, and the woman «extended her hands and blessed to heaven Monheim which took place in proximity to the saint’s relics, with a narrative of her him: he who heals and strikes down, the Lord» (ibidem). One senses in Wolfhard’s lived virtues: chastity, humility, obedience, religious fervor, and missionary activity. account a tone of wonder, and also an anxiety to remind his reader that it was not In the first three antiphons, the poet relates how Walpurga enthusiastically took Saint Walburga, and certainly not the prioress of Monheim, who were responsible up monastic discipline and a chaste life, and was consumed with the ardent flame for the miracles: rather, he repeatedly insists, the pilgrims were healed by their of the divine Christ. In the next three antiphons, he describes the only miracles to faith in Christ, the only one capable of granting miracles. Jane Tibbetts Schulenberg have taken place during Walburga’s lifetime: the healing of a mortally ill girl, and (2008, p. 76) has remarked on the attempt of the Carolingian-era church to control the appearance of a miraculous light emanating from the saint in prayer. The text increasingly popular saints’ cults and pilgrimage sites. Schulenberg writes that of antiphon 2.3 emphasizes her desire to avoid attention for these miracles: «These «well aware of the authority, power, prestige, and economic advantage associated [miracles] having been seen, she was moved by neither human praise nor favor, with saints and pilgrimage centers, bishops attempted to monitor these saints’ cults but truly served with humility in her breast». Antiphons 3.1 and 3.2 explain that and the whole area of the production of miracles». The quickly growing pilgrimage Walburga should be celebrated instead for her asceticism and accomplishments as site at Monheim posed a problem for the episcopal establishment at Eichstätt for a missionary: «Not without cause is this virgin be praised — she, who tested herself several reasons. First, the fame of Monheim had eclipsed that of the older shrine like gold in the furnace. For behold, for God, she prepared for them in darkness the at Walburga’s tomb in Eichstätt, which had more established ties to the cathedral. eternal light of the Lord». These statements praising Walburga’s humility, asceticism, Second, as Schulenberg has shown, religious women directly administered access and missionary work over the miraculous presents an implicit critique of the rising to the saint’s tomb and mingled with pilgrims of both sexes in the healing rituals of power of the Monheim convent, which had garnered fame and wealth through its Monheim, in violation of the 816 Institutio sanctimonalium that required canonesses association with Walburga’s wonder-working relics. By contextualizing the recent to live in complete claustration (ibidem, pp. 76-80). The healing shrine at Monheim miracles of Monheim against the model of female sanctity described in the vita, furthermore offered an alternative path to sanctity, one achieved neither through the office domesticates the miraculous and refocuses attention on a more orthodox the sacraments nor through the singing of the office, but rather through proximity understanding of Christianity. to Walburga’s relics. Thus the leadership of Monheim enjoyed a position of spiritual authority that was unmediated by the male clergy. Wolfhard’s account also reveals In addition to Trier 5, the office for Saint Walburga is transmitted in adiastematic notation that the sound of the convent chapel was of a very different nature, one less orthodox in the fragment D-Mü Clm 29316/5 (antiphoner, Salzburg, tenth-eleventh century) and and far less regulated than the antiphons, psalms, and hymns that marked the ritual in the twelfth-century notated breviary CZ-Pu VI.E.4c from the convent of St. George of Herrieden and Eichstätt. Seen in this context, Wolfhard’s office can be read as a in Prague. Portions of the office are found in diastematic notation in three sources: mirror that reflects the wonder of the miracles of Monheim, while simultaneously three matins responsories in NL-ZUa 6, the lauds antiphons in CZ-Pu XIV.B.13 and the reinforcing an orthodox image of female monasticism as recorded in the vita. Magnificat antiphon for second vespers in the printed Antiphonale Monasteriense of 1537 (cantusdatabase.uwaterloo.ca). The three responsories transmitted in NL-ZUa 6 The prose texts of the office for Walburga were adapted from Wolfhard’s vita and clearly share a contour with the neumed versions in Trier 5, albeit with some melodic miracle books into a tightly organized secular cursus. While most saints’ offices simplification. The responsories exemplify two distinct musical styles: the first is a bridge between the historical time of the saint’s life and prayers that take place in the conventional and formulaic Romano-Frankish style, exemplified by responsories 1.1 present day, the office for Walburga is unusual in that it has three layers of time to Haec sollemnis est dies and 3.3 Sancta Walpurgis Christi. The second, exemplified by negotiate: the distant past of the vita, the more recent past of the posthumous miracles, responsory 1.2 Beatissimae Christi virginis, is a freer and more dramatic style that and finally, the present in which the office is performed. The office is organized in displays early tendencies of the so-called ‘middle’ or ‘transitional’ period of office

416 417 CANTUS PLANUS 2014 ALTSTATT — SINGING THE SAINTS IN MEDIEVAL EICHSTÄTT composition, and which contains instances of parallelism and text painting.3 These Example 2. Attributed to Wolfhard of Herrieden, R. 1.2 Beatissimae Christi virginis (NL-ZUa 6) two distinct compositional styles complement the content and function of each text.

The opening responsory 1.1 Haec sollemnis est dies is an invitation to prayer in the here and now. As shown in example 1, the melody of this first mode responsory is utterly formulaic.

Example 1. Attributed to Wolfhard of Herrieden, R. Haec sollemnis est dies, opening (NL-ZUa 6) and anonymous, R. Descendit de caelis, opening; ed. Walter H. Frere (1966)

Beatissimae Christi virginis relates a miracle that took place during the translation to Example 1 compares Haec sollemnis est dies with the first mode responsory Descendit Monheim: that of a tiny boy who was knocked down by Walburga’s bier and trampled de caelis, which Walter Howard Frere (1966, p. 22). identified among the oldest and in the procession. When his lifeless body was again put in contact with the bier, he most conventional group of first mode responsories The two responsories follow a miraculously regained his health. The first thing to notice about this responsory is its similar melodic contour and the same series of cadences. In comparison with the sheer immensity. While the typical eighth mode responsories transcribed by Frere typical first mode responsories transcribed by Frere, Haec sollemnis est dies appears to contain five-seven phrases comprised of 110-151 pitches, Beatissime Christi virginis be of average length and phrase structure. The conventionality of its music suits the contains nine phrases that total 165 pitches (Frere 1966, p. 53). The structure and style and function of its formulaic prayer text. ordering of its phrases are also unusual. While it is not unheard of for certain eighth mode responsories to begin with a phrase which cadences to D, it is certainly not This musical style contrasts with that of the miracle stories: a freer and more dramatic typical. This responsory is unusual in that it begins with a set of three such phrases style that incorporates elements of parallelism and text painting. Example 2, transcribed that together create a melismatic meditation at the bottom of the modal range, that from NL-ZUa 6, gives Responsory 1.2, the eighth mode Beatissimae Christi virginis, the slowly invokes the name: Beatissimae Christi virginis Walburga. The next clause is first of the set of responsories that narrates the miracles of Monheim. Notably, the text set with a pair of equally unusual parallel phrases, each of which ends in rhyming mentions neither Monheim nor its founder Liubila. melismatic cadence to c, near the top of the modal range. One is tempted to hear in this pair of precadential melismas that waver between c and d Walburga’s bier lifted 3. For a discussion of stylistic periodization of the medieval proper office, see Goudesenne (2004); Hankeln (2001); Hiley high, making its bumpy journey to Monheim. The next phrase arcs down from c to (2001 and 2004); Dobszay and Szendrei (1999); Crocker (1986).

418 419 CANTUS PLANUS 2014 ALTSTATT — SINGING THE SAINTS IN MEDIEVAL EICHSTÄTT

G on the word parvulus or «tiny», as if to underscore the child’s small stature. In the The Magnificat antiphon for second vespers is an early example of the so-called next phrase, the melody sinks to low D on caducus as the child falls. It then recites on ‘double antiphon’— a form that repeats a received melody twice or more in the setting the final of G before resolving to F, the subfinal, and most unstable place in the mode, of a single text. As László Dobszay and Janka Szendrei have described (1999, p. 156), mirroring the uncertainty of his predicament. Parallelism, now at the motivic level, the melody may be repeated exactly, extended, or abbreviated, creating longer, marks a repeated figure on the word et – the turning point of the entire story – before multi-part parallel forms inflected with subtle variants and ornamental elaboration the melody surges up through a triadic motive on meruit – «deserved» – to once (Altstatt 2011, pp. 438-493 and 2013, pp.187-190). We can conceive of Pretiosa et again achieve the c reciting center. The last word sospitatem or «safety» – makes optima dei as modeled on a simple first mode melody that consists of three phrases a leisurely descent to the final cadence on G. The unusually long verse Ac ille that that cadence to D, C, and again to D. In Pretiosa et optima dei, the melody has been follows (not shown) adapts a conventional two-part responsorial tone by repeated extended by the insertion of a second initial cadence on D, resulting in a four- a variation of the a section, resulting in an aa1b form similar to that used by introit phrase form. The text is set to two elaborate variations on this extended melodic verses (Frere 1966, p. 4). form. The second iteration of the melody skips the opening ‘a’ phrase entirely. Found throughout the tenth-century office, including examples attributed to Stephen of The responsories of the third nocturn return to prayers in the present day: the only Liège and to Reginold of Eichstätt, the double antiphon is a musical-poetic genre that transmitted example of this set, the responsory Sancta Walpurgis Christi, returns merits closer study as a hallmark of the so-called ‘middle’ or ‘transitional’ period to the same conventional musical style of Responsory 1.1 Haec sollemnis est dies of office composition. The Walburga office’s use of the double antiphon, along with described above. Thus the musical styles complement the function of each text: the its innovative approach to received responsory melodies, exemplifies the creative freer, more dramatic style is used to set the miracle stories of the recent past, in adaptation of received melodic models that were used alongside new styles in office contrast to the conventional style used to set formulaic prayers uttered in the here composition from the turn of the tenth century. and now. Finally, the layout of the late tenth-century manuscript Trier 5 offers important clues as to how its scribe may have viewed the Walburga office in relationship to the later Example 3. Attributed to Wolfhard of Herrieden, Magnificat antiphon Pretiosa et optima dei, Antiphonale offices for Willibald and Wunnebald. The three offices were copied together into Monasteriense, Köln: Hero Alopecius (Fuchs), 1537. nineteen bifolios, folded to create a libellus with a blank cover — a small book of 16,4 cm by 11,4 cm, which could have easily been held in the cantor’s hand. This cantor’s libellus was only later bound together with the rest of the hagiographic material that comprises Trier 5. It is noteworthy that the scribe of the libellus included the matins lessons for the offices for Willibald and Wunnebald,but did not do so for the office for Walburga. This suggests that her office was already established in the repertoire, with the lessons held in a different liturgical book, whereas those for her brothers were new additions to the liturgy. Furthermore, the Walburga office is the only one of the three to be copied in order of liturgical performance. The offices for Willibald and Wunnebald, by contrast, list matins antiphons and responsories in separate sets by genre, and not in order of liturgical performance. This again suggests that the Walburga office was probably copied from an older liturgical source, while the offices for Willibald and Wunnebald were new, and presented in such a way as to emphasize the cyclic composition of their matins responsories and antiphons. Finally, the three offices do not appear in the order of the liturgical calendar, but rather are listed according to the gender of the saint and their relative importance the Eichstätt cathedral: first Willibald, Eichstätt’s celebrated founding bishop, then Wunnebald, the first abbot of Heidenheim, and finally Walburga, abbess and celebrated wonder-worker, now demoted in favor of her two brothers. These details of layout support the notion that the scribe, and possibly Reginold himself, conceived of the offices for Willibald and Wunebald as the

420 421 CANTUS PLANUS 2014 ALTSTATT — SINGING THE SAINTS IN MEDIEVAL EICHSTÄTT concluding members of a trilogy for the Eichstätt patrons: one that re-ordered the three century composition, and the role of the sounding proper office in defining sacred siblings in a new hierarchy of importance in relation to the cathedral, and established history, in reordering relationships between institutions, and in establishing models Reginold within a creative lineage that began with the Eichstätt poet and composer of orthodoxy. Wolfhard of Herrieden. Indeed, Anonymus Haserensis’s description of the Feast of St. Willibald at the newly expanded cathedral recalls Wolfhard’s account of the miracles at Monheim. Quoting from Reginold’s lauds antiphons for Willibald, he relates how, at the time of the saint’s translation in 989: Bibliography

all signs and wonders had gradually ceased, where previously there had been many. And as many Antiphonale Monasteriense (1537), Köln: Hero Alopecius (Fuchs). of us have often heard told, every year when the Feast of St. Willibald drew near, such a multitude of the blind, deaf, mute, and those possessed by demons and various afflictions gathered that the Altstatt Alison (2011), The Music and Liturgy of Kloster Preetz: Anna von Buchwald’s Buch im Chor in its fifteenth-century Context, Ph. D. dissertation, University of Oregon. great colonnade at St. Martin could not hold them all. And they made such a racket on the holy vigil that it interfered not a little with that most joyful vespers. Nevertheless, before mass on the – (2013), ‘The Rhenish Heritage of the Preetz Antiphoner: Origins and Reform’, Journal of the Alamire holy day, it ended, nearly all having been healed and giving thanks to God and to Saint Willibald. Foundation 5, pp. 175-199. And upon this, that same bishop [Reginold] composed the [lauds] antiphon Surdos, caecos, mutos, mancos, and the previous one that ends with miraculis signorumque variis virtutibus prepotentem – (2014), ‘Review: Digital and Multimedia Scholarship. Reviewed Works: Cantus Planus Regensburg by David Hiley; Corpus Antiphonalium Officii-Ecclesiae Centralis Europae by László Dobszay, Gábor effecerat (Weinfurter 1987, pp. 48-49). Prószéky; CANTUS: A Database for Latin Ecclesiastical Chant by Debra Lacoste, Jan Koláček, Kate Helsen; Global Chant Database by Jan Koláček; The CANTUS Index by Jan Koláček, Debra Lacoste, The presence of the Walburga office in Trier 5 tells us that it was part of the liturgy of Elsa De Luca, Kate Helsen’, Jams 67/1, pp. 267-285. the Eichstätt cathedral from the tenth century. The transmission of its texts as late as – (2015), D-TRb Abt. 95, Nr. 5 in Cantus: a Database for Latin Ecclesiastical Chant, cantus.uwaterloo.ca. 1560 demonstrates that the office remained in the cathedral’s repertoire for at least six hundred and fifty years (Breviarii iuxta ritum Ecclesiae Aichstadiensis, Pars Hyemalis Bauch Andreas (1979), Ein bayerisches Mirakelbuch aus der Karolingerzeit, Regensburg: Pustet (Quellen zur Geschichte der Diözese Eichstätt 2; Eichstätter Studien Neue Folge 12). 1560, fols. 313-315v). It is possible that the cathedral shared this office with the canonesses who resided at the nearby Eichstätt tomb of Saint Walburga. Unfortunately, Breviarii iuxta ritum Ecclesiae Aichstadiensis, Pars Hyemali (1560), Ingolstadt: n.pr. this is impossible to determine, as the community, which adopted the Benedictine Rule Crocker Richard (1986), ‘Matins Antiphons at St. Denis’, Jams 39, pp. 441-490. in 1035, possesses no manuscripts for the office from before the seventeenth century (Lechner 1937, pp. 17-19). However, the office’s appearance in the breviary CZ-Pu Dobszay, László - Szendrei Janka eds. (1999), Antiphonen, Mmmae 5.

VI.E.4c and the antiphonary CZ-Pu XIV.B.13 from the Saint George cloister in Prague Dörr Friedrich - Schlager Karlheinz - Wohnhaas Theodor (1987), ‘Spicilegia Willibaldina: Musikalische hints at a possible transmission through a female Benedictine lineage.4 This, together und literarische Gaben zu Ehren des Eichstätter Bistumspatrons aus mittelalterlichen Quellen’, with the office’s emphasis on Walburga’s virtuous life, suggests the possibility that it Studien und Mitteilungen zur Geschichte des Benediktiner- Ordens und seiner Zweige 98, pp. 36-62. may have been written with the Eichstätt convent in mind, offering simultaneously an Frere Walter Howard ed. (1966), Antiphonale Sarisburiense: A Reproduction in Facsimile of a Manuscript of orthodox presentation of her life and miracles, and a model of female monasticism. the Thirteenth Century, with a Dissertation and Analytical Index, Farnborough: Gregg.

Goudesenne Jean-François (2004), ‘A Typology of Historiae in West Francia (8-10 c.),’ Pmm 13, pp. 1-31. In conclusion, I have argued that the office for Saint Walburga contained in Trier 5 is likely the same one attributed to Wolfhard of Herrieden by Anonymus Haserensis. Hankeln Roman (2001), ‘Properization and Formal Changes in High Medieval Saints’ Offices: the Offices for Saints Henry and Kunigunde of Bamberg’, Pmm 10, pp. 3-22. Part of Wolfhard’s effort to explain Walburga’s burgeoning cult at Monheim, the office can be seen as a dual effortto legitimize the cult by aligning it with sanctioned Hiley David (2001), ‘The Music of Prose Offices in Honour of English Saints,’P mm 10, pp. 23-37. liturgical celebration, and to promulgate an orthodox model of female sanctity – one – (2004), Early Cycles of Office Chants for the Feast of Mary Magdalene in John Haines and Randall which posed an implicit critique of the un-orthodox soundscape and compromised Rosenfeld eds., Music and Medieval Manuscripts: Paleography and Performance. Essays Dedicated to claustration of Monheim. Finally, it offers a case study in the techniques of early tenth- Andrew Hughes, Burlington: Ashgate, pp. 369-400.

Hoffmann-Brandt Helma (1971), Die Tropen zu den Responsorien des Officiums,Ph. D. dissertation, Friedrich-Alexander-Universität, Erlangen-Nürnberg. 4. The transmission of the office for Saint Walburga in the fragment D-Mü Clm 29316/5 (tenth-eleventh century, Salzburg) and that for Saint Willibald in the fragment A-Smi 492 (eleventh century, Salzburg) point to possible transmission to Prague through Salzburg. Imma, first Abbess of the Benedictine Saint Walburga Abbey, had previously been a nun at the convent of Holzbauer Hermann (1972), Mittelalterliche Heiligenverehrung: Heilige Walpurgis, Kevelaer: Butzon and Nonnberg in Salzburg (Zunker 2009, p. 13). A continued relationship between the two houses is not implausible. Bercker (Eichstätter Studien Neue Folge 5).

422 423 CANTUS PLANUS 2014

Michelle Urberg Jones Charles W. (1963), The Saint Nicholas Liturgy and Its Literary Relationships (Ninth to Twelfth Centuries), Berkeley: University of California Press. Learning gender roles Lechner Joseph (1937), Die spätmittelalterliche Handschriftgeschichte der Benediktinerinnenabtei St. Walburg/ Eichstätt (By.), Münster: Verlag der Aschendorffschen Verlagsbuchhandlung. from the Paschal and Mariological processions of the brothers and sisters at Vadstena Abbey Schulenberg Jane Tibbetts (2008), Women’s Monasteries and Sacred Space: The Promotion of Saints’ Cults and Miracles in Lisa M. Bitel - Felice Lifshitz eds., Gender and Christianity in Medieval Europe: New Perspectives, Philadelphia: University of Pennsylvania Press, pp. 76-80.

Talbot Charles H. ed. (1954) The Anglo-Saxon Missionaries in Germany, Being the Lives of SS. Willibrord, Hiis vero diebus debent sorores circuire ambitus suos cantando quod inter se statuerint, una Boniface, Leoba and Lebuin Together with the Hodoepericon of St. Willibald and a Selection from the sorore crucem ante eas ferente, scilicet: die resurrectionis Domini et diebus rogationum, Correspondence of St. Boniface, London-New York: Sheed and Ward. ascensionis, penthecostes et quinque solennitatum beate virginis scilicet concepcionis, nativitatis, annunciacionis, purificacionis et assumpcionis, et in natali apostolorum Petri et Wendehorst Alfred (2006), Das Bistum Eichstätt 1: Die Bischofsreihe bis 1535, Berlin-New York: de Gruyter 1 (Germania Sacra Neue Folge 45). Pauli, et in die sancti Michahelis et omnium sanctorum. Et fratres similiter faciant. Indeed, on these days, the sisters should walk around their walkway while singing that which Weinfurter Stefan ed. (1987), Die Geschichte der Eichstätter Bischöfe des Anonymus Haserensis. Edition. they decided amongst themselves, one sister bearing the cross before them, namely on: the Übersetzung. Kommentar, Regensburg: Pustet (Eichstätter Studien Neue Folge 24). day of the Resurrection of the Lord, the days of Rogation, the Ascension, Pentecost, and the five solemnities of the Blessed Virgin – the Conception, Nativity, Annunciation, Purification, and – (2011), «Überall unsere Heiligste Mutter Walburga»: Entstehung, Wirkkraft und Mythos eines europäischen Heiligenkults in Gert Melville and Anne Müller eds., The Female «Vita Religiosa» Assumption – and the day of Apostles Peter and Paul, the day of Saint Michael, and All Saints. between Late Antiquity and the High Middle Ages: Structures, Developments, and Spatial Contexts, And the brothers should do the same. Berlin-Münster-Wien-Zürich-London: Lit (Vita regularis: Ordnungen und Deutungen religiosen Lebens im Mittelalter Abhandlungen 47), pp. 187-206. When the first Birgittine brothers and sisters took orders on 23 October 1384, at Wood Susan (2006), The Proprietary Church in the Medieval West, Oxford-New York: Oxford University Vadstena Abbey, they began to fulfill what Birgitta envisioned for the liturgical Press. practice of her Order. The brothers sang offices and masses according to the

Zunker Maria Magdalena (2009), Geschichte der Benediktinerinnenabtei St. Walburg in Eichstätt von 1035 Linköping usage and the sisters sang the seven weekly Marian rhymed offices bis heute, Lindenberg: Fink. – called the Cantus Sororum – and daily mass Salve sancta parens (Gejrot 1996, no. 41; Eklund 1975, p. 118; Hollman 1956, p. 129). These liturgies were entirely independent and were to be performed serially every day in the cloister church, where the brothers stood in their choir at the western end and where the sisters stood in their balcony at the eastern end. The Birgittine brothers would begin singing the liturgy proper to the day and hour, then the sisters would follow with their liturgy for the day and hour. The church was intentionally designed so that each group would hear the other singing. Their daily liturgies were, however, only a portion of their antiphonally-organized liturgical practice. Even though Birgitta prescribed the daily liturgies, as the above quotation from Prior Petrus Olavi’s additions to the Birgittine Rule (Additiones prioris Petri ad regulam sancti Salvatoris, hereafter Additiones) suggests, specific feast days were identified to supplement their liturgical practices with processions.2 It is hard to say if both the brothers and the sisters had fleshed out an initial processional repertory by the time that Petrus Olavi had prepared the list of required feast days in the Additiones, because almost no manuscript evidence

1. Latin from Nyberg (1972-1974, p. 107). I thank Volker Schier for providing helpful suggestions with some of the Latin translations. Unless otherwise indicated, translations in this article are my own.

2. The five Marian feasts Petrus lists represent a modification from the Birgittine Rule, which identifies the Purification (2 February), the Annunciation (25 March), the Assumption (15 August), and the Nativity of Mary (6 September) as feast days (Eklund 1975, p. 110).

424 425 CANTUS PLANUS 2014 URBERG — LEARNING GENDER ROLES FROM THE PROCESSIONS AT VADSTENA ABBEY remains from the earliest decades at Vadstena. It is certain, nevertheless, based on the last quarter of the fifteenth century.5 In addition, two ordinals – S-Uu C 428 and manuscript evidence, that they did develop their feast day processional repertories S-Sr Skoklostersamlingen No. 2 E 8899 (henceforth Skokloster 2) – provide a clearer and continued to revise them throughout the fifteenth and early-sixteenth centuries.3 perspective on how the brothers’ repertory changed over the course of the fifteenth Extant manuscripts containing these processions indicate that the brothers and the century. Skokloster 2 dates from the end of the fourteenth century and records an sisters cultivated different repertories and continued to modify them over the years. older usage of the Linköping Diocese than is documented in S-Uu C 428, a mid- Moreover, at least some of the revisions of chant texts and melodies shed light on the fifteenth century document by brother Laurentius Törnevalla (likely at Vadstena symbolic functions of the chant for the brothers and sisters, respectively. This study between 1448-1458).6 The processional repertories of the sisters are recorded in an explores a few of these changes in the processional antiphons and responsories ordinal and in a variety of processionals and choir directories. The sources come for feasts that comprise the heart of the Birgittine processional repertory and that from a similar period as the brothers’ manuscripts: most of the processionals and reflect the importance of Christological and Mariological devotion in the Birgittine choir directories date to around the turn of the sixteenth century (S-Uu C 470, C Order, namely those for Good Friday and Easter and those for the Marian feasts 458, C 472, C 473, and C 508), with only the ordinal (Lampeter, Wales, UK, Saint of the Purification and Visitation.4 The gendered processional repertories of these David’s University College Library, MS. 5, henceforth Lp 5) and one choir directory feasts reflect and reinforce the social roles of the brothers and sisters in the cloister. (S-Uu C 442) witnessing to an earliest period, namely, from the middle part of the fifteenth century. These manuscripts, placed in conversation with the available The purpose of this study is at least two-fold. First, it illustrates, for the first time, evidence from the additions quoted at the beginning of this article, offer snapshots part of the Birgittine processional repertories, which have received comparatively little of salient moments in the devotional life at Vadstena. Evidence in these manuscripts attention from chant scholars, outside of the processions of the Birgittine sisters at demonstrates that the processional repertory is more flexible than other parts of the Syon, discussed briefly by Anne Bagnall Yardley in Performing Piety (2006), and those brothers’ diocesan liturgy or the Cantus Sororum of the sisters. Examples contained of the Birgittine sisters at Maihingen, discussed by Volker Schier and Corinne Schleif, in several manuscripts used by the brothers and one by the sisters demonstrate points in Katerina’s Windows (2008) and in Schier’s recent article on the Birgittine nuns’ of change over the course of the fifteenth and early sixteenth centuries. An example processions at Maihingen (2013). This project examines the processions of the brothers in the brothers’ processional repertory comes from their chants for the Feast of the and the sisters at Vadstena, thereby expanding the previous discussions of Birgittine Purification (2 February). The responsory Responsum accepit Symeon and its verse processional chant. Second, it contributes to our understanding of the symbolism of Cum inducerent were part of the diocesan repertory according to both Skokloster chant more generally. The symbolic meanings of the Birgittine processional repertory 2 and S-Uu C 428,7 and they were also originally in S-Uu C 506 and C 481, which are related to their gendered status: particular chants have been assigned exclusively suggests they were sung at least as early as the fourteenth century in the Linköping to the brothers or exclusively to the sisters. Although this division of liturgical practices Diocese and continued to be used at Vadstena until the turn of the sixteenth century. along gender lines is unique to the Birgittine Order, motivations of the brothers and At some point in the last decade of the fifteenth century, however, these chants were sisters for choosing particular chants can speak to repertorial choices across western monastic traditions. As we will see, it will be possible to glean a deeper understanding 5. Schier (2013, insert) has grouped S-Uu C 506 among the sisters’ processional manuscripts. The textual scripts, notation, and repertory suggest that the sisters were the original scribes for this manuscript, but that the brothers of the symbolic meanings of feast day processional chants through a closer examination appropriated it and modified it to suit their liturgical needs. The dimensions of S-Uu C 506 may also corroborate that of the Birgittine processional repertories. it was made for the brothers at 8.9 inches tall and 6.3 inches wide (22.5×16 cm). The sisters’ choir directories and processionals range in height between 5.5 and 7.3 inches (14-18.5 cm) and in width between 4.3 and 5.1 inches (11-13 cm). The brothers’ manuscripts, in comparison, vary from 6.3 to 8.7 inches tall (16-22 cm) and from 4.5 to 6.3 inches wide (11.5-16 cm). On the whole they are one to two inches taller and wider than the sisters’ manuscripts, suggesting The chant repertories that I have consulted for this study are recorded in antiphoners, that, based on size, C 506 was intended for the brothers even though it was possibly made in the sisters’ cloister. processionals, choir directories, ordinals, and customaries. Most come from the 6. Like many brothers at Vadstena, Laurentius wrote sermon manuscripts, though he is particularly known for being a Uppsala University Library C-Collection (S-Uu C), which has the largest repository compiler of sermon texts rather than a creator (Borgehammar 1995, pp. 97-98). of music manuscripts used at Vadstena Abbey (see table 1, p. 434). The brothers’ 7. Skokloster 2 (N viii): «Ante processionem sacerdos benedicit cereos cum oracione Benedic domine cum ceteris et prefacione. Deinde aspergantur aqua benedicta et thure adoleantur et accendantur. Interim cantetur antiphona processional repertories can be found in S-Uu C 481 and C 506, two manuscripts from Lumen ad reuelacionem psalmus Nunc dimittis cum suis versibus Sequitur oracio Omnipotens semper deus. Quia finita processio cum hiis antiphonis Aue gracia plena, alia Adorna thalamum, antiphona Responsum acceperat Intrando ecclesiam dicantur antiphone Cum inducerent alia Hodie beata virgo Sequitur V. Post partum, coll. Domine Ihesu Christe. Ad missam». 3. Several different versions of the Additiones are known to have existed. The earliest one, according to Henrik Vitalis, S-Uu C 428 (fols. 67v-68): «Ante quod processio fiat sacerdos benedicat cereos cum oracione Benedic domine et ceteris dates to 1384 and another important one to 1420. These two are in Latin. Another was translated to Swedish for the sisters Cum prefacione Deinde aspergantur aqua benedicta et thure adoleantur et accendantur Interim cantetur antiphona during the fifteenth century (Vitalis 1995, pp. 51-52). My article concerns itself with the version from 1384. Lumen ad reuelacionem psalmus Nunc dimittis cum suis versibus ac ad quemlibet versum repetatur antiphon Et cum Gloria patri Sequitur oratio Omnipotens semper Quia finita fiat processio cum hiis antiphon Ave gracia plena alia 4. In addition to the feast days listed in the opening quotation, the Birgittine Order celebrates the Visitation of Mary (31 Adorna thalamum, alia Responsum Intrando ecclesiam dicantur antiphona Cum inducerent alia Hodie beata virgo May), the Presentation of Mary (21 November), and the Conception of Mary (8 December). Sequitur V. Post partum coll. Domine ihesu christe Ad missam».

426 427 CANTUS PLANUS 2014 URBERG — LEARNING GENDER ROLES FROM THE PROCESSIONS AT VADSTENA ABBEY removed from the repertory, as evidenced in S-Uu C 506 (fol. 3) where Responsum example, the brothers at Vadstena commemorated Christ’s death and resurrection accepit Symeon and Cum inducerent have been scraped away from the parchment with several sets of processions to and from the sepulcher for the Depositio, Elevatio, and replaced with the Pentecost antiphon Hodie completi sunt. In S-Uu C 481 (fols. Visitatio sepulchri, and Aspersio aquae.11 They began with the Depositio crucis 40v-41), it is simply marked vacat because the paper did not allow for new material veneration, singing the responsory Ecce quomodo moritur and verse In pace while to be written over it.8 carrying the cross on the way to the sepulcher, followed by the antiphons Caro mea and In pace factus est at the deposition of the cross, and concluding with the responsory An example of change and flexibility in the sisters’ processional repertory can be seen Sepulto domino and verse In pace as they returned to the choir. During the Easter vigil, with the revisions made to chant texts for the feast of the Visitation of Mary (31 May) they processed to the hymn Inuentor rutili. On Easter Sunday, they performed a large and two chants: the responsory O ineffabiliter divitem and the antiphon Magnificetur. number of processions in honor of the Resurrection: the antiphons Surrexit dominus, The Visitation is not listed in the opening quotation from the Additiones, almost Cum surgit, and Cum rex glorie before the Elevatio, the responsory Ego sum alpha and certainly because this feast was not officially added to the calendar at Vadstena until verse Ego sum vestra at the Elevatio, and the antiphons Vidi aquam, Salve festa dies, after 1389 (Schmid 1938, p.152; Helander 1957, p. 42).9 By the third quarter of the Sedit angelus, Crucifixum in carne, Nolite metuere dico, and Recordamini quomodo at fifteenth century, however, it had become a codified part of the repertory. These chants the Aspersio aquae. By comparison, the sisters venerated Easter with processions on are absent entirely from the choir directories dated to the middle and the second third Good Friday and Easter Sunday that generally reflected their commitment to following of the fifteenth century (S-Uu C 442, C 483, and C 485), but they appear faithfully in the weekly themes of the Cantus Sororum.12 The Friday procession is woven into the rubrics for the Visitation in later choir directories and processionals, along with a third unique narrative of the Cantus Sororum as a weekly commemoration of both Christ’s antiphon Exultant sancti (S-Uu C 508, C 470, C 458, C 472, and C 473). The ordinal (Lp 5, suffering and death and Mary’s co-suffering with Christ. The sisters did not process fol. 46v) records O ineffabiliter and Magnificetur for the feast of the Visitation of Mary.10 specifically for the Depositio, but instead would conduct their usual Friday procession Several notated manuscripts dating to the turn of the sixteenth century and after (S-Uu to the sepulcher in the cloister church at Terce, as well as the procession in which they C 468, C 472, C 473, and C 458) record them as processional chants for the Annunciation, read the penitential psalms, as specified in the additions.13 On Easter Sunday, they a feast that had long been celebrated with the responsory Quomodo fiet istud and verse processed only before and during the Elevatio, singing the antiphons Surrexit dominus, Ideoque quo, and with the antiphon Hec est dies. This manuscript evidence suggests Cum rex glorie, and Christus resurgens, the responsory Ego sum alpha and verse Ego that the feast of the Visitation was codified before the turn of the sixteenth century. It sum vestra, and concluded with Regina celi, but they did not process for the Aspersio seems also that the themes associated with the Visitation were eventually assigned aquae. Their singing of Regina celi as an Easter procession further emphasized the to the Annunciation. Regardless of the reasons that prompted these changes to their themes they celebrated weekly in the Cantus Sororum, reflecting Marian themes they processional repertory, these chants show how the Birgittine sisters continued to re- emphasized in celebrating even non-Marian feasts.14 imagine what they should sing on feast days throughout the fifteenth century. Whatever decisions were made to change the processions, the overall devotional function of the 11. The processionals S-Uu C 506 (fols. 9v-11v) and S-Uu C 481 (fols. 46v-50) contain the processional chants for the brothers’ and the sisters’ chant repertories remained stable throughout the fifteenth venerations that happen between Good Friday and vespers on Easter Sunday. These ceremonies were part of the Linköping and sixteenth centuries: they reinforced a gender divide in the liturgies practiced at tradition long before the Birgittine brothers took them up in their ceremonies (Davidson 1990). Vadstena. We can see this in the striking gender-based assignment of chants for the 12. Two sets of themes unfold daily in the Cantus Sororum, which originated with the twenty-one matins readings revealed to Birgitta (the Sermo Angelicus): one theme pertains to salvation history (creation through the resurrection of Christ), and Paschal season. On the days of Good Friday, Holy Saturday, and Easter Sunday, for the other to Mary’s participation in salvation history: Sunday explores the creation, joy in the triune God, and Mary as the ideal model for creation. On Monday, the beauty and fall of the angels is paired with Mary being venerated by the angels. Tuesday continues with the fall of Adam, the role of the Old Testament patriarchs, and Mary as the premeditated protector of fallen humanity. Wednesday studies the birth of Mary, her childhood, and the Immaculate Conception. Thursday brings 8. It is notable that the same scribe – Michael Nicolai – corrected both C 506 and C 481 at some point during his career Mary’s life into conversation with Christ’s through texts and music about the incarnation and the annunciation. Friday’s at Vadstena between 1487 and 1516. Michael Nicolai was a scribe at Vadstena who may have specialized in liturgical office venerates Christ’s suffering and death and Mary’s co-suffering with Christ. Saturday concludes the week, with the manuscripts. His hand is also seen in two collections of rhymed offices S-Uu C 23 and C 21. celebration of the Virgin’s faith in Christ, as well as her own death and assumption into heaven.

9. Schmid (1938) and Helander (1957) rely on a deed of gift dated 2 July, 1389, in which Erengisle Sunesson gives land in 13. The Friday procession to the sepulcher was codified in the Rule for the Birgittine Order, but described in more detail Lundby to the nuns’ cloister at Skänninge (not far from Vadstena). This letter identifies the day of composition as the feast in the Reuelaciones Extravagantes and in the Additiones (Eklund 1975, p. 110; Hollman 1956, p. 122). The Additiones has of the Visitation (S-Sr SDHK-nr: 13545). The Visitation is present in the Linköping Diocese around the turn of the fifteenth the most detailed description: «Qualibet sexta feria debent sorores circuire ambitus suos insimul legendo septem psalmos; century, as evidenced by the rubric in Skokloster 2 Qiii: «In festo visitacionis beate virginis usque ad ocatuis apostolorum quibus finitis intrabunt chorum suum et precedende abbatissa vel alia sorore ad hoc deputata flexis genibus legant letanias» inclusiue fiat memoria de apostolis». The Vadstena diary also contains an entry that acknowledges the official acceptance (Nyberg 1972-1974, p. 107). of the Visitation in the Linköping Diocese. On 2 July 1406, the Synod decided that the Feast of the Visitation of Mary should be celebrated after the octave of John the Baptist as a double feast, but not a land feast («Festum visitacionis beate Virginis 14. Regina celi is a somewhat special case because it is one of the four Marian antiphons that cycle every year according celebretur in crastino octavarum Iohannis Baptiste sub festo duplici, sed non sub festo terre», Gejrot 1996, no. 146). to the liturgical calendar. Even though this chant supports my argument about inserting chants about the Virgin on non- Marian feast days, it is the Marian antiphon that is normally assigned to the time between Holy Saturday and the octave of 10. Lp 5 primarily represents the repertory used at Vadstena ca. 1450, but certain feasts were added for the creation of Lp 5 Pentecost. A particularly Birgittine example of interpolating a Marian antiphon into a the context of a non-Marian feast is in 1481. The Visitation is one of those feasts. the singing of Ecce Maria for Midsomar (the summer solstice).

428 429 CANTUS PLANUS 2014 URBERG — LEARNING GENDER ROLES FROM THE PROCESSIONS AT VADSTENA ABBEY

Although the Eastertide processions serve to demonstrate how the Birgittines assigned Stella sub nube tegitur maria mundo premitur rutilans in splendore Elyzabeth perducitur ad chants to their processional repertories based on gender, through an analysis of the solamen lux spargitur roborans in vigore. V. Luna soli coniungitur Elyzabeth deuoluitur estuans chant texts we will see the character of the gendered identities in their liturgies. The texts in amore. Gloria patri et filio et spiritui sancto. 18 chosen for the feast of the Visitation of Mary illustrate this argument. For this feast, the The star is covered, behind the cloud; Mary is overwhelmed by the world, reddening in splendor. Elizabeth is guided to the source of solace; the light is scattered, strengthening in vigor. V. The sisters sang the responsory O ineffabiliter and verse Hic ad patriam, and the antiphons moon is united with the sun; Elizabeth falls in love. Magnificetur and Exultant sancti. In the responsory O ineffabiliter and its verse Hic ad patriam, the sisters rejoice for Mary’s motherhood and extol her queenship: The antiphon Ihesu redemptor optime, their other processional chant for this feast, is a prayer to Christ directing Mary and Elizabeth to bestow blessing on those celebrating O ineffabiliter divitem in paupere opido puellulam Que missum sibi superni regis filium suscepit in aulam iocunditatis desideratis pavit deliciis, et armis potencie circumcinxit. V. Hic ad patriam the Visitation: triumphator rediens in perhenni laude ipsam reginam constituit.15 O ineffably wealthy maiden in a poor town! The Son of the Highest King who was sent to her, she Ihesu redemptor optime ad mariam nos imprime vt mundi aduocata pari forma nos visitet sicut received into [her] hall of joyfulness. She fed [him] with delightful pleasures and girded him with fecit elyzabeth per summam pietatem mores et actus dirigat et ad celos alliciat per gratiam 19 weapons of power. V. Here the triumphant one, when he returns to the fatherland, builds up the collatam. very Queen [Mary] with everlasting praise. Jesus, most excellent redeemer, imprint Mary on us, so that she, the world’s advocate, may visit us in the same form as she visited Elizabeth. May she direct our habits and acts through supreme The antiphons Magnificetur rex and Exultant sancti are similarly laudatory, though in piety to heaven through the conferral of grace. the former, the admiration for Mary is couched in Christological terms: It is no surprise then, that Stella sub nube / Luna soli and Ihesu redemptor optime Magnificetur rex celestis milicie qui ancille sue humilitatem respiciens super omnia que creavit are, in fact, traditionally associated with the Visitation.20 As suggested through this ipsam magnificauit laudetur incessanter eius benigna caritas qui ipsam nobis matrem constituit example of the feast of the Visitation, texts of the processional chants of the brothers’ caritatis.16 and sisters’ repertories have different topical foci. The brothers’ chants focus primarily Let the king of the celestial army be glorified. For he, when he gazed on the humility of the on the biblical themes of the feast day, while the sisters’ chants are addressed to Mary handmaiden, glorified her above all things that he created. Let his kind charity be praised regardless of the feast-day biblical narrative. unceasingly, he has created for us the very mother of charity.

I argue that the Birgittines crafted the gendered identities of their processional texts to In the latter, the text praises Mary’s holiness: reinforce particular social roles in the cloister, namely the brothers as preachers and Exultant sancti de Maria sanctificacionis archa que ex gracie quam apud deum inuenit plenitudine sisters as contemplatives. The origin of the gender divide in the Birgittine processional omnium ad se fugiencium depellit inopiam. 17 repertory comes, in part, from the Birgittine Rule, in which the brothers devote The saints praise the holiness of Mary; the ark from the grace of which, in the presence of God, themselves to preaching and the sisters promise themselves first and foremost to came the plenitude of all men; she dislodged the need of their fleeing. contemplating the Virgin Mary (Eklund 1975, pp. 107-108, 121; Hollman 1956, pp. 116, 129, 133).21 We can see how the Birgittines reinforced these identities, in part, through The textual themes of the sisters’ processions for the Visitation focus primarily on the sisters’ weekly liturgy. Their liturgical life is purposefully meditative in that they extolling Mary rather than on narrating the theme commonly associated with this return every seven days to the same music and texts in the Cantus Sororum, whose feast: Mary’s visit to Elizabeth when Mary is pregnant with Christ and Elizabeth is texts variously employed either newly composed or borrowed from existing material. pregnant with John the Baptist. The brothers’ processional chants for the feast of the The origins of the Cantus Sororum texts are particularly important here. Thematically, Visitation, in comparison to those of the sisters, are not moments to meditate on the they are based on the twenty-one matins readings of the Sermo Angelicus, but consist virtues of the Virgin Mary, but instead they describe the biblical themes associated with this feast. The responsory Stella sub nube and its verse Luna soli, for example, 18. S-Uu C 481, fol. 57r-v; C 506 fols. 31v-32. acclaim the heavenly response to Mary’s and Elizabeth’s meeting: 19. S-Uu C 481, fols. 57v-58; C 506, fol. 32.

20. See Cantus Id nos. 602252, 602252a, and 202738.

15. Latin from manuscript D-FS Alto MS P AN 2. 21. See chapter 15, which clarifies that the brothers should preach «Christus precipit hic, quod fratres sacerdotes dicti 16. Latin from Servatius 1990, pp. 186-187. monasterii debeant tantummodo oracioni et leccioni vacare et certis diebus euangelium in materna lingua ipsis monialibus predicare», and chapter 1, which identifies the sisters as the raison d’être of the Order «Hanc igitur religionem ad honorem 17. Latin based on Servatius 1990, p. 188. amantissime Matris mee per mulieres primum et principaliter statuere volo».

430 431 CANTUS PLANUS 2014 URBERG — LEARNING GENDER ROLES FROM THE PROCESSIONS AT VADSTENA ABBEY of musical and textual material that is newly composed to accommodate Marian This brief survey offers an introduction to the Birgittine processional repertories themes unique to the Birgittine Order or borrowed from existing Marian feast day sung at Vadstena that have heretofore been discussed only in regard to their basic chants (Lundén 1976; Eklund 1972).22 The sisters’ processions thus complement their content. Manuscript evidence suggests that the brothers’ and sisters’ processional contemplative identity because they are drawn largely from the Cantus Sororum, and repertories continued to develop on gendered trajectories throughout the life of the in the case of the feast of the Visitation, the chants borrowed are among those that were monastery between 1384 and about 1525. Whatever repertorial changes were made, newly composed to suit the exclusively Birgittine ideas in the Monday and Tuesday Birgittine processional chants convey a message about the devotional life at Vadstena offices. O ineffabiliter / Hic ad patriam is, for example, borrowed from the third that is connected to the specific roles of the brothers and sisters. A close look at responsory for matins on Tuesday. In the Cantus Sororum, this responsory is paired several Paschal and Mariological processional chants reveals that the gendered with a reading that describes Mary as the premeditated protector of the fallen. The repertories, when considered alongside the particular roles of brothers and sisters as other two processional antiphons for the Visitation were written for the Monday office identified by the Rule, actually reinforce the idea that the brothers be preachers and of the Cantus Sororum: Magnificeturis sung before the Magnificat and Exultant sancti the sisters be contemplatives. Birgittine processional chants suggest that feast day is sung during compline. The themes explored in the matins readings of Monday’s celebrations can accrue broader meanings within an Order that go beyond simply office focus on the creation of the angels and subsequent fall of Lucifer, how the angels elevating a saint or an important ecclesiastical event. Processional chants can shed knew Mary would be created, that she was with God and the angels before she took light on devotional identities of those pouring forth the song in the ambulatories of on her earthly body, and that humanity is instilled with free will so that they might do the cloister church. virtuous deeds. By choosing chants that they would sing weekly for their feast days, the Birgittine sisters intentionally transferred the Cantus Sororum’s Marian themes to their processions. As a result, their feast day processional chants for the Visitation commemorate Mary primarily as the premeditated protector, which is a theme touched on only secondarily in the brothers’ processions (e.g. in Ihesu redemptor, in which Mary is ‘the advocate for the world’).

That the brothers’ processional chants are a function of their role as teachers in the Order is best understood in the broader context of the output of their scriptorium. Most of the brothers’ extant codices contain sermons or theological texts rather than liturgical music. The two processional manuscripts discussed here (S-Uu C 481 and C 506) are part of a very small set of extant notated liturgical manuscripts containing the brothers’ offices and masses.23 Even if their library was known to be complete, it would almost certainly contain fewer liturgical manuscripts than the library of the sisters because they focused primarily on sermon writing. This extant set of liturgical manuscripts can, as Emma Dillon (2011, p. 392) argues, function as «the physical remains of lost sound [that] encode in their very materiality a little of the values and meanings of music in the world beyond their bindings». In other words, their processionals convey the same values as their sermon manuscripts: they are musical expansions of their preaching and teaching. Seen in this light, the brothers’ feast day chants Stella sub nube / Luna soli, and Ihesu redemptor optime are ‘proper’ to their role in the monastery as sermonizers. These processions for the Visitation function as aural extensions of their mission to preach Mary and Elizabeth’s blessed pregnancies.

22. See note 12 for the summary of these themes.

23. Other notated liturgical manuscripts extant today include rhymed offices (S-Uu C 21 and C 23) and antiphoners (S-Uu C 426 and C 450).

432 433 CANTUS PLANUS 2014 URBERG — LEARNING GENDER ROLES FROM THE PROCESSIONS AT VADSTENA ABBEY

Table 1. Manuscripts consulted for this project Bibliography

Borgehammar Stephan (1995), Preaching to Pilgrims: Ad vincula Sermons at Vadstena Abbey in Monica Manuscript Sigla Genre Date Scribe Who used it Hedlund ed., A Catalogue and Its Users: A Symposium on the Uppsala C Collection of Medieval (Short titles) at Vadstena Manuscripts, Uppsala: Acta Universitatis Upsaliensis (Acta Bibliothecae R. Universitatis Upsaliensis Dombibliothek Freising Cantus Sororum Fifteenth century Sisters 34), pp. 91-100. D-FS Alto MS P An 2 Antiphonal Eklund Sten ed. (1972), Sancta Birgitta Opera minora II: Sermo Angelicus, Stockholm: Almqvist & Wiksell (MS P An 2) (Samlingar utgivna av Svenska Fornskriftsällskapet 2. Latinska skrifter band 8:2). Stockholm, Riksarkivet, Ordinal Late-fourteenth Brothers Skoklostersamlingen, (Linköping Usage) century – ed. (1975), Opera Minora I: Regula salvatoris, Lund: Berlingska boktryckeriet (Samlingar utgivna av No. 2 E 8899 Svenska Fornskriftsällskapet 2. Latinska skrifter band 8.1), (Skokloster 2) Davidson Audrey Ekdahl (1990), Holy Week and Easter Ceremonies and Dramas from Medieval Sweden S-Uu C 428 Ordinal ca. 1441-1458 Laurentius Törnevalla Brothers Kalamazoo, MI: Medieval Institute Publications, Western Michigan University. (Linköping Usage) S-Uu C 481 Processional Late-fifteenth Michael Nicolai Brothers Dillon Emma (2011). Music Manuscripts in Mark Everist ed., The Cambridge Companion to Medieval century (fols. 2, 36-39v, 40, Music, Cambridge: Cambridge University Press, pp. 291-319. 41v-76) Gejrot Claes ed. (1996), Vadstenadiariet: Latinsk text med översättning och kommentar, Stockholm: Gotab S-Uu C 506 Processional Before 1508 Michael Nicolai Brothers (Kungliga Samfundet för utgivande av handskrifter rörande Skandinaviens Historia 19). (fols. 1, 22v, 23, 29, 32v, 33v, 34-36v) Helander Sven (1957), Ordinarius Lincopensis c:a 1400 och dess liturgiska förebilder, Lund: Cwk Gleerup.

Lampeter, Wales, UK, Ordinal 1481 (copy of a Christina Hansdotter Sisters Hollman Lennart ed. (1956), Reuelaciones extrauagantes, Uppsala: Almqvist & Wiksell (Samlingar utgivna Saint David’s University ca. 1450 version Brask av Svenska Fornskriftsällskapet 2. Latinska skrifter band 5). College Library, of the sisters’ MS. 5 (Lp 5) ordinal) Lundén Tryggve ed. (1976), Officium parvum beate Marie Virginis,Uppsala: Uppsala University Press (Acta S-Uu C 442 Choir Directory fifteenth century Sisters Universitatis Upsaliensis. Studia historico-ecclesiastica Upsaliensia 27-28).

S-Uu C 483 Choir Directory Mid- to Late- Sisters Nyberg Tore (1972-1974), Dokumente und Untersuchungen zur inneren Geschichte der drei Birgittenklöster fifteenth century Bayerns, 1420-1570, 2 vols., Munich: Beck.

S-Uu C 485 Choir Directory Mid- to Late- Sisters Schier Volker (2013), Writing and Rewriting Processions: The Reworking of the Procession Liturgy by the fifteenth century Birgittine Nuns of South Germany in Claes Gejrot - Mia Åkestam - Roger Andersson eds., The Birgittine Experience: Papers from the Birgitta Conference in Stockholm 2011, Kvhaa konferenser 82, S-Uu C 468 Choir Directory Late- fifteenth Sisters Stockholm: Kungl. Viterhets Historie och Antikvitets Akademien, pp. 268-287 and insert. century S-Uu C 508 Choir Directory 1473-1486 Christina Hansdotter Sisters Schier Volker - Schleif Corine (2008), Katerina’s Windows: Donation and Devotion, art and music, as heard Brask and seen through the writings of a Birgittine Nun, University Park, PA: Pennsylvania State University.

S-Uu C 470 Choir Directory Turn of sixteenth Probably Christina Sisters Schmid Toni (1938), ‘Franziskanische Elemente im mittelalterlichen Kult Schwedens. Schluß’, century Hansdotter Brask Franziskanische Studien 25, pp. 134-161. S-Uu C 458 Choir Directory 1518 Anna Germundsdotter Sisters Servatius Viveca (1990), Cantus sororum: musik- und liturgiegeschichtliche Studien zu den Antiphonen des birgittinischen Eigenrepertoires, Stockholm: Almqvist & Wiksell. S-Uu C 472 Processional Turn of sixteenth Christina Hansdotter Sisters century Brask Vitalis Henrik (1995), Addiciones Prioris Petri. On Brigittine Rules and Regulations in Monica Hedlund S-Uu C 473 Processional 1501-1519 Katarina Gudmundi, Sisters ed., A Catalogue and Its Users: A Symposium on the Uppsala C Collection of Medieval Manuscripts, Anna Fickesdotter Uppsala: Acta Universitatis Upsaliensis (Acta Bibliothecae R. Universitatis Upsaliensis 34), pp. 47-56.

Vouri Hilkka-Liisa (2011), In the Herb Gardens of Oure Ladye: The Great Responsories of Matins in Bridgettine Sisters’ Liturgy of Hours, Ph. D. dissertation, Sibelius University, Helsinki.

Yardley Ann Bagnall (2006), Performing Piety: Musical Culture in Medieval English Nunneries, New York: Palgrave Macmillan.

434 435 CANTUS PLANUS 2014

Manuel Pedro Ferreira Sitography

Svenskt Diplomatariums huvudkartotek (Sdhk) Venetian influence in fifteenth-century Portugal http://sok.riksarkivet.se/SDHK

Sdhk-nr: 13545 https://sok.riksarkivet.se/bildvisning/Sdhk_13545_13545.jpg https://sok.riksarkivet.se/sdhk?SDHK=13545&postid=sdhk_13545

Solange Corbin identified and explored long ago a striking connection between Portugal and the Veneto, mainly in her magisterial book La déposition liturgique au Vendredi Saint.1 More recently, Lucia Moro with Giulio Cattin (1993) and Barbara Haggh (2009) revisited the issue. This connection involves the presence in both Portugal and the Veneto of the same planctus followed by a series of versicles, sometimes described as Lamentation tropes, sung during the Good Friday procession for the burial of the Lord. The text, Heu heu Domine, heu heu Salvator noster, is found with its notation in only three Italian sources, from Padua and Venice, from either the second half of the fifteenth century or the early sixteenth (table 1).2 They are all polyphonic two-voice versions; Solange Corbin was aware of only one, marked with an asterisk in the table.

Table 1. Italian notated sources of the planctus Heu heu Domine.

*1.a Padova, Biblioteca capitolare nella curia vescovile, I-Pc C 56, cathedral processional, addition from the second half of the fifteenth century, fol. 61. Free composition 1.b Venice (1506 print), Lamentationum Jeremie prophete: same contents as I-Pc C 56; attribution to Johannes de Quadris, active in San Marco in the mid-fifteenth century 2. Padova, Biblioteca del Seminario, I-Ps ms. 359, (ritual from San Giorgio in Alga, Venice, 1505), fol. 71v. Recitational tenor (G final)

In contrast, the Portuguese musical sources, none of them earlier than the first quarter of the sixteenth century, are plentiful. The corresponding tradition eventually reached Brazil, India, and Japan. Practical written sources were produced well into the twentieth-century (Azevedo 1960, pp. 64-68, being the latest); oral performance survives to this day. In most places that keep the procession for the burial of the Lord, the planctus is no longer sung; however, it can still be heard in continental Portugal (Braga; Monsanto), the Azores islands (Vila de Povoação, S. Miguel) and Brazil (São

Manuel Pedro Ferreira, Centro de Estudos de Sociologia e Estética Musical, Nova Fcsh, Lisboa.

1. Corbin 1947; Corbin 1952, pp. 302-15; Corbin 1960. The contribution of Corbin for the study of musical sources in Portugal is explored in Ferreira (2011, 2015). On Corbin (1960), see also Palazzo (2015).

2. A later, hypothetical connection with the Veneto is signaled by Corbin (1960, p. 153): the singing of O vos omnes, «venu tout droit des cérémonies de Padoue et Venise», by a female singer taking the role of Veronica, attested in Lisbon and Brazil in the nineteenth century; the custom was widespread but was eventually abandoned in most places, although it is still found in popular tradition today, e.g., in Livramento (Mafra), Vila Boim (Elvas), Vila de Povoação (Azores) and many Brazilian towns.

436 437 CANTUS PLANUS 2014 FERREIRA — VENETIAN INFLUENCE IN FIFTEENTH-CENTURY PORTUGAL

3 João del-Rei, Minas Gerais), where surviving scores date from the late eighteenth and Monophonic versions (in print) nineteenth centuries (Corbin 1960, pp. 156-158; Castagna 2001; Fonseca 2003). Table 2 lists the Portuguese continental sources, written up to ca. 1900, thus far identified. *1. Lyon (1558), Missale iuxta usum et ordinem Almae Bracarensis Ecclesiae Hispaniarum Primatis, fol. 96 (G final) They are organized by type, date, and melodic version: when two or more sources coincide and clearly derive from the same tradition, they are counted as a single *2.a Lisboa (1607), Liber processionum, et stationum Ecclesiae Olysiponensis v version.4 Solange Corbin was aware of only ten notated sources of the planctus, one of fol. 73 (E final) which was Italian (rather than three) and the other nine Portuguese (rather than forty- *2.b Lisboa (1728), Liber processionum, et stationum Sanctæ Metropolitanæ Ecclesiæ three)5. These are marked in the tables with an asterisk. Ulyssiponensis Orientalis, p. 52 (E final) 3.a Lisboa (1610), Processionarium Fratrum Ordinis Virginis Mariæ de Monte Carmelo Table 2. Portuguese notated sources of the Heu heu planctus (to ca. 1900) p. 68 (E final)

Monophonic versions (in manuscript) 3.b Lisboa (1642), Processionarium Fratrum Ordinis Virginis Mariæ de Monte Carmelo p. 68 (E final) 1.a Braga, Arquivo da Sé, P-BRs, sixteenth-century Cathedral processional [A] fols. 109v-111 (G final) 4.a Coimbra (1620), Processionarium monasticum iuxta consuetudinem Monachorum Nigrorum Ordinis S. P. N. Benedicti Regnorum Portugaliæ 1.b Braga, Arquivo da Sé, P-BRs, sixteenth-century Cathedral processional [B] pp. 122-123 (G final) fols. 102-103: same contents 4.b Coimbra (1691), Processionarium monasticum […], pp. 82-83 (G final) 1.c Braga, Arquivo Distrital, P-BRad Ms. 646, sixteenth-century Cathedral processional) same contents *4.c Coimbra (1727), Processionarium monasticum […], p. 90 (G final) 1.d Braga, Arquivo Distrital, P-BRad Ms. 726, Directório e cerimonial do Coro da Sé seventeenth-century copy, fols. 52-53 2. Vila Viçosa, Paço Ducal, Biblioteca D. Manuel II, P-VV Res. Ms. 85 Adq. 5. Lisboa (1626), Manuale Chori secundum usum fratrum minorum, et monialium Ceremonial of the Royal Chapel [1554-1557], fol. 35 (E final) S. Claræ, pp. 93-94 (E final) 3. Chicago, Newberry Library, US-Cn MS 155, Portuguese processional 6.a Coimbra (1675), Fr. Raymundo da Converçam, Manual de tudo o que se canta fora sixteenth-century addition, fol. 80v (E final) do choro conforme ao uso dos Religiosos, e Religiosas da sagrada ordem de Penitencia do nosso Seraphico Padre São Francisco do Reyno de Portugal, pp. 243-244 (E final) 4.a Évora, Sé, P-EVc cód. perg. lit. 8, sixteenth-century Cathedral processional fols. 44-45 (E final) 6.b Lisboa (1780) Processionale, ac rituale romano-seraphicum ad usum fratrum, ac monialium S. Francisci, pp. 36-38 (same contents as above) 4.b Évora, Sé, P-EVc cód. perg. lit. 9, sixteenth-century Cathedral processional same contents *6.c Lisboa (1803) Processionale ac rituale romanum cum officio sepulturae 5. Évora, Sé, P-EVc, no call number, truncated College processional dated 1657 pp. 54-55 (same contents as above) fol. 23 (E final) 6.d Lisboa (1826) Processionale, ac rituale romano-seraphicum ad usum fratrum, 6. Leiria, Arquivo Distrital, P-LE, Regras de contraponto, sixteenth-century addition ac monialium S. Francisci, pp. 48-50 (same contents as above) fols. 75-77 (G final) 7. Lisboa (1730), Manual dos religiosos da SS. Trindade, e redempçam de cativos deste Reyno de Portugal, p. 65 (E final) 3. Braga maintains monophonic performance by two (high-pitched) singers of a planctus whose melody was, however, possibly restored in the early twentieth century, in accordance with the 1924 missal; Monsanto has a modified melody, trusted 8. Lisboa (1733), Manuale processionum, cum benedictionibus solemnibus […] to the Three Maries (http://casadasbestas.blogspot.pt/2011/04/d.html); Povoação (video recording, 2012: http://www.youtube. collectum opera, et industria fratrum Ordinis Eremitarum, ac Descalceatorum com/watch?v=A0lIG3hB-ew, at 7’50’’) has simple, two-voice male polyphony (the missing traditional melody may have been modified over time or entrusted to a higher voice); S. João del-Rei (LP record, 1977, http://www.youtube.com/watch?v=Fo__ S. Augustini in Regno Portugalliæ, p. 106 (E final) LwWSg28, at 7’ 30’’) uses two-voice female polyphony of a more classical stamp, freely derived from a G version of the planctus. *9 n.p.; n.d., Directorium chori ad usum fratrum excalceatorum S.P.N. Augustini 4. Sources that present the versicles only are excluded, although these may impart significant information, e.g. P-AR Res. Ms. 32 (polyphonic choirbook, ca. 1615), fols. 96v-98, the earliest source connected to a Cistercian environment, where the Portugaliae, p. 81 (E final) G final in the cantus implies a compatible planctus melody. The listing in table 2 incorporates, among other sources, those cited by Corbin (1952; 1960) and data kindly provided by João Pedro d’Alvarenga (personal communication, 27/07/2014). 10. Lisboa (1788), Bernardo da Conceição, O ecclesiastico instruido scientificamente na arte do canto-chão, pp. 448-451 (new melody, E final) 5. Corbin (1952, p. 310), in the midst of her presentation of Reis 1837, gives a modified melody (starting like the E final melody, but clearly centred on A with G final); this book, however, does not include musical notation. I was unable to trace any source with the version transcribed by Corbin. Since she mentions a copy of the Lisbon 1728 processional in the organ 11. Porto (1801), Manuale chori ad usum canonicorum praeclarae congregationis tribune at the cathedral in Lamego, one may speculate that the melody was an addition to, or modification of, books kept in S. Joannis Evangelistae, pp. 46-48 (E final) Lamego, but this hypothesis would have to be investigated.

438 439 CANTUS PLANUS 2014 FERREIRA — VENETIAN INFLUENCE IN FIFTEENTH-CENTURY PORTUGAL

Polyphonic versions G; those connected to the south (the Royal Chapel, Lisboa, Évora, the Order of Saint James, the Carmelites, the Trinitarians and fringes of the Cistercian and Franciscan ca. 1600-1700 Orders) adhere to the melody ending on E. After 1750, modified or free melodies appear *1. Coimbra, P-Cug M.M. 53 (ca. 1600), early addition on fols. 124v-125, four voices; in parallel with the traditional melody, and the fourth heu is frequently dropped. The Cantus: original G final melody, modified at the end to arrive on E earliest mention of a polyphonic performance of the Heus dates from the Royal Chapel 2. Porto, P-Pm MM 76, early seventeenth-century Cantus partbook, Ceremonial of ca. 1555 and implies a four-part texture: «those who are to say the verses p. 119, four voices, E final of the Heu heu will start it in unison from a booklet that shall be given to them, and 3. Porto, P-Pm MM 76, early seventeenth-century Cantus partbook, at the end of each verse four or eight singers will respond in polyphony, Heu heu».6 p. 138, four voices, G final Likewise, the seventeenth-century Carmelite Processional, whose melody coincides 4. Évora, Sé, P-EVc, no call number, truncated College processional dated 1657, fol. 27v, precisely with the one sung in the Royal Chapel (although without the last heu), seems four voices, Cantus: G final to echo this practice, specifying «quatuor qui Heu, Heu, cantu organico modulabuntur». 1701-1800 At that time, the verses themselves could also be set for several voices. In surviving polyphonic manuscripts (starting ca. 1600) A-G and F-E versions of the melody could *5. Lisboa (1757 print), Cistercian processional, p. 59, four voices, Cantus: E final, coexist in the same source or group of sources, and both versions reached Brazil. transposed

6. n.p., Lisboa, Biblioteca Nacional de Portugal, P-Ln MM 3208//1-8, four voices, Only two pieces are found in the three Italian sources. The first is a two-voice free melody freely recomposed composition by Johannes de Quadris, active in Venice in the mid-fifteenth century. 7. n.p., Lisboa, Biblioteca Nacional de Portugal, P-Ln MM 1871//1-5v, four voices, The second is an unrelated two-voice version, contained in the ritual from the canons Tiple: modified traditional melody regular of San Giorgio in Alga, a tiny, now-abandoned island in the Venetian lagoon, 8. n.p., Lisboa, Biblioteca Nacional de Portugal, P-Ln, MM 2282//1-4, four voices, one kilometre west of the Giudecca (vaporetto no. 16 passes close by on its way to the Soprano: ornamented traditional melody, G final Fusina Terminal). This was a reformed community inspired by ascetic ideals, whose 9. n. p., Lisboa, Biblioteca Nacional de Portugal, P-Ln MM 4776//1-4, four voices, statutes were granted in 1404 and which drew its ranks from local nobility. During Tiple: melody transposed to the E final the fifteenth century, the congregation expanded into many towns, including Venice 10. n.p., Lisboa, Biblioteca Nacional de Portugal, P-Ln MM 4769//1-3, incomplete: SAT; itself, where the old Abbey of the Madonna dell’Orto (now known as the burial site of B missing; free composition Tintoretto) was given to the canons in 1468 (Tomasini 1642; Cornaro 1758; Gallicciolli 11. n.p. [female institution], Lisboa, Biblioteca Nacional de Portugal, P-Ln MM 223//1-4: 1795; Cracco 1959). SA + instrumental B; Soprano: G final

12. n.p. [female institution], Lisboa, Biblioteca Nacional de Portugal, P-Ln MM 3275//1-3, Example 1. Versions of the planctus Heu, heu Domine: San Giorgio in Alga and Braga incomplete: SA; B missing; Soprano: E final 13. Óbidos (eighteenth century), Igreja de S. Pedro, n/no. 9 (not seen by author) I-Ps, ms. 359 1801-ca. 1900

*14. Coimbra, P-Cug M.M. 112 (m. 16), an 1825 copy of an earlier manuscript, music attributed to Toscano, early seventeenth century, four voices, Cantus: G final P-BRsc 15. Guimarães, Arquivo Municipal Alfredo Pimenta, P-G C-922, no. 70: nineteenth- or early twentieth-century parts, four voices, Soprano: G final.

The Heu heu melody is notated in the Portuguese sources either around the A and 6. P-VV Res. Ms. 85 Adq., Ceremonial da capella del Rey [D. João III], fol. 35: «os que hão de dizer os versos do heu, heu, o começarão unisonus em hum caderno que lhes sera dado, e no cabo de cada verso quoatro ou oyto cantores risponderão ending on G, or around F, with E final. A consistent pattern seems to emerge for the de camto dorgão heu heu». The ceremonial includes two versions of the Burial ritual for Good Friday (fols. 25-35v): the period 1550-1750: sources connected to northern, or predominantly northern areas first ignores the processional planctus; the second includes it. This small manuscript, eventually used in Oporto cathedral as a notebook in the late sixteenth century, was recently acquired by the House of the Braganzas, and has so far escaped (Coimbra, Braga, Guimarães, and the Benedictines) adhere to the melody ending on scholarly attention. On the Royal Chapel practices, see also Andrade (1653, pp. 111-123).

440 441 CANTUS PLANUS 2014 FERREIRA — VENETIAN INFLUENCE IN FIFTEENTH-CENTURY PORTUGAL

The recitational nature of the tenor in the San Giorgio ritual, centred on A with its the burial procession with its chanting was a novelty introduced to the country in the final on G, is a strong indication that it derives from an earlier plainchant tone. This second half of the fifteenth century. is confirmed by its similarity with the Portuguese melody (example 1). Note that the psalmodic use of a reciting tone a whole tone or a semitone above the final was found The path by which the burial rite and its planctus reached Portugal is in need of in both the Hispanic and the Gallican-Ambrosian traditions (Ferreira 2006). clarification. Abbot João Álvares mentions the rite in his 1467 letter to the Benedictine monks of Paço de Sousa (Oporto).10 Nonetheless, it is reported that, not long afterwards, Solange Corbin (1960, pp. 23, 141-143), followed by Moro and Cattin, proposed that the canons of Saint John the Evangelist (who were also known as Lóios) independently the Portuguese imported the planctus with its traditional plainchant melody from adopted it in imitation of or inspired by a journey to Jerusalem, and that they were the Veneto, possibly Padua. Recently, Barbara Haggh proposed (2009, pp. 21-23) an responsible for its dissemination. To reconcile these facts, Corbin tentatively suggested alternative hypothesis that the planctus travelled instead from Braga in Portugal to that the planctus was introduced into Portugal twice in the second half of the fifteenth 11 the Veneto. She makes the entirely reasonable assumption that the Heu, heu, Domine century. trope existed in Braga before its earliest surviving sources; that Humbertus de Salinis, Reference to Jerusalem relates to the fact that a well-known canon, Paulo de Portalegre a singer and composer known to have been granted a canonicate at the cathedral of (1437-1510), visited the Holy Land around 1470, and, sometime during his life, wrote Braga before 1403, adopted the trope for polyphonic composition; and that the idea a directorium describing the burial ceremony. A rumour connecting this with the was possibly transmitted to Johannes Ciconia during the Council of Pisa of 1409- origins of the practice could have been spread to enhance the ceremony’s liturgical 1410, where both may have met, reaching Padua, and then Venice, in due course. authority. Portuguese historiography is unanimous in attributing the introduction of This hypothesis is interesting but not readily acceptable. Humbertus de Salinis was the planctus to the Lóios canons, yet the most plausible scenario, as argued below, most likely hired by the Portuguese court in Lisbon, where the archbishop of Braga, is that they adopted it from Venice as early as 1453, before it reached Paço de Sousa Martim Afonso, said to have been a former ambassador to France and one of the main making it unnecessary to suppose two successive imports. The Order of the Canons counsellors to King John I, had been in residence since the end of 1401.7 Humbertus of Saint John the Evangelist was founded in Lisbon around 1420 by Master Joham probably remained with the Portuguese court for over six years, since, by the time he Vicente (1380-1463), physician for the Prince Regent Dom Duarte. Irresistibly attracted became a papal singer in 1409 (having travelled with the archbishop of Lisbon João and immersed in Dominican liturgy from an early age, it was only at the behest of his Afonso to attend the Council of Pisa), he had been promised additional benefices at parents’ that Master John refrained from entering the Order at that time. He took up Coimbra and Évora when they became vacant, which implies that he had supporters the idea again when he reached middle age, but ultimately he decided along with a at the highest political level.8 A polyphonic fragment survives in Lisbon from this group of courtiers to create a clerical community free of perpetual vows. After a move period (Ferreira 2008, 1, p. 49, and plates XIII-XIV). The probability that Humbertus to Oporto in 1423 and the following dispersal, in 1425 they established themselves heard the planctus in Lisbon is quite slim. In a description of the Good Friday ritual at at Vilar dos Frades (Barcelos, diocese of Braga). After an initial period of uncertainty the Portuguese court from around 1435, written for the Prince Regent Dom Duarte, no concerning the forms of canonical organization, in 1427 the Order approved the model processional planctus is mentioned.9 Furthermore, Portuguese chroniclers agree that of San Georgio in Alga in Venice, renowned for its ideals of poverty, simplicity and zeal in the pursuit of ecclesiastical reform, although Augustinian rule was still considered 7. On Archbishop Martim Afonso Pires da Charneca (Martinho V, 1398-1416), originally from Lisbon, formerly a married a viable alternative until the confirmation in 1431 of Venetian-based constitutions man and bishop of Coimbra (1392-1398), see Cunha (1634-1635, 2, pp. 216-222); Morais (1943-1948, t. III/1, pp. 140-41); Matos (2009, pp. 423-38); Soveral (2004-2014, addendum for I, p. 128, l. 43). In the late fourteenth century, Humbertus, to (Portalegre 2007, pp. 79-88, 193-196; Santa Maria 1697, pp. 555-556; Tavares 2001). judge from his ballade En la saison, was connected to or at the service of the Du Guesclin family; given its proximity to the French royalty, he may have met Martim Afonso during the latter’s sojourn in France. That same year Joham Vicente was appointed bishop of Lamego by the pope; he would

8. The papal letters connecting Humbertus de Salinis to Portugal were uncovered by John Nádas at the Vatican Secret continue as head of the Order, but at a distance. Archive: RL 111 (Boniface ix, 29 May 1403: anno 14, lib. 158), fol. 182v, and RL 138 (Alexander v, 10 July 1409: anno 1, Lib. 5), fols. 105-106. There the composer, a deacon, is described as holding a canonicate with prebend at Braga; he had previously exchanged an additional benefice from the rents of São Pedro de Torrados (Felgueiras) for another from São At this point, there were only ten canons. The initial group had been joined in 1430 by Salvador de Tagilde (Vizela), both, at the time, in the diocese of Braga. The first letter allows him to add more benefices to the chapel-master of the Infant D. Fernando, Joham d’Arruda, who had been reared these two, worth together eighty Tours pounds. The second letter reserves for him, already a papal singer, a canonicate at the cathedral of Lisbon worth eighty Tours pounds, previously held by D. Fernando da Guerra, just nominated bishop of Silves, on the condition that he relinquishes expectations of prebends at Coimbra (worth fifthy Tours pounds) and in Évora. The letters and the corresponding summary descriptions were kindly provided to me by John Nádas (personal 10. Leão de Santo Tomás 1974, pp. 265, [29]; Ribeiro 1810, 1, pp. 356-63 [360]; Calado 1993. communication, 23/9/2014). See also Nádas (2007). 11. Corbin 1952, pp. 309, 314-315; Corbin 1960, pp. 154-155. The introduction of the ceremony is attributed to Paulo de 9. The text is found in two sources: Duarte (1982, pp. 215-17; 1998, pp. 346-48). Paraphrase and analysis in Ferreira 2014. Portalegre by Santa Maria (1697, pp. 300-305, 863). Corbin supposes that his journey to Jerusalem took place late in his life. The fact that the Royal Chapel Ceremonial of ca. 1555 admits two forms of the burial rite suggests that the second form, the For correct dating, see Portalegre (2007, ‘Introdução’ and p. 134). Another author attributing to Lóios the introduction of the one with the processional planctus, had been introduced to the court not long before. ceremony, besides the four quoted in Corbin (1960, pp. 261, 268, 271 and 274), is Santos (1734, p. 534).

442 443 CANTUS PLANUS 2014 FERREIRA — VENETIAN INFLUENCE IN FIFTEENTH-CENTURY PORTUGAL in Lisbon’s diocese, and led Vilar dos Frades between 1434-1439, 1441-1443 and 1449-1451. Under his guidance both the ceremony and the singing were reformed, and stabilized, possibly with the help of Vasco Gonçalves, formerly a chapel singer in the service of the Duke of Braganza, who joined the congregation and served as prior of Vilar dos Frades between 1446-1448 and again in 1458 (Pina 2011). Interest in polyphony among the canons (indirectly borne out by the possession of books of motets issued by Pierre Attaignant in 1534-1535, still extant in Lisbon), may have started at an early stage with Joham de Arruda and Vasco Gonçalves. It is said that Arruda brought from Venice, where he sojourned for some time between 1452 and 1453 on his return from Rome,12 the ordering of liturgy and some rules of chant; a late testimony attributes to him the authorship of treatises on music and ritual.13 When he arrived in Portugal a prose, Omnis caetus clericorum, was newly composed and sung in his honour during the washing of the feet, by his friend Baptista (who had accompanied him to Rome but returned earlier). The prose was later kept in the Order to accompany the washing of the feet of occasional clerical guests.14 In San Giorgio, in identical circumstances, a lauda by Leonardo Giustiniani was sung, notated in the same manuscript as the planctus. This practice may have inspired canon Baptista.15 We lack any detailed information about any other liturgical practice in the Order before the sixteenth century.16 Later in the century the Portuguese canons were called back to Venice to reform the community of San Giorgio, which in the meanwhile had fallen into disrepute. The connection with the canons of San Giorgio in Alga is significant here. Moro and Cattin established that their liturgy was related to the reformed Benedictines of the congregation of Santa Giustina, and that their music encompassed simple forms

12. João de Arruda travelled to Rome with a priest called Baptista to defend the congregation from claims advanced by Dom Fernando da Guerra. Baptista appears in the records only from 1449 onwards, and probably died in 1558; a critical period in papal acknowledgement of the congregation rights can be placed at around 1551-1553; but Arruda was rector of Vilar dos Frades until 1451, so the journey could not have taken place until 1552.

13. Portalegre 2007, pp. 152-153: «trouxe a emformação dos modos e cirimonias dos Irmãos de Italia […] Este foi ho que nos deu regra do modo de rezar e cantar he ordenou hos tõs segundo acostumamos e asi as outras sanctas cerimonias». The account of João d’Arruda by Santa Maria (1697, pp. 741-754), is not always trustworthy, but confirms his role in shaping the Order’s liturgy: «Compoz varios tratados, assim de solfa, como das ceremonias […] não só servio, mas illustrou a Congregação: porque dispoz com grande primor, & méstria, a forma de cantar, & celebrar o Officio Divino […] Trouxe de lá [São Jorge de Alga] algumas regras conducentes à mayor perfeição do canto, & do culto divino, que entre nós se admittirão, & praticárão». The Infant D. Fernando owned a missal of the use of Lisbon, as seen in his will (P-Lant GAV-16-2-13), but was well acquainted with the Sarum rite, on account of both his spiritual interests and his royal English blood (his mother, the Queen, was a Lancaster): cf. Álvares (1730, pp. 33-38).

14. A notated version was recently found in a miscellaneous codex (breviary-ritual-antiphoner), P-EVp Cod. 524 (Fundo da Manizola), fol. 75. The manuscript was used at the Convento de Nossa Senhora da Assunção at Arraiolos, founded in 1527.

15. Portalegre 2007, pp. 174, 180-181; Santa Maria 1697, p. 731; Moro and Cattin 1993, pp. 180-181. The lauda is attributed to Giustiniani in Wilson (2009).

16. Only a few liturgical manuscripts have been identified so far. The earliest seems to beP-EVp Cod. 524, referred to above (n. 14), whose original layer probably dates from the late fifteenth century or ca. 1500. The second volume of a sixteenth-century gradual at the Biblioteca Nacional (P-Ln L.C.10) is largely a palimpsest made to conform to Roman usage; my attempt to recover the primitive layer with the help of technicians at Instituto José de Figueiredo did not lead to significant results. Later sources include P-Ln L.C.9, a psaltery-hymnary written shortly after 1631 (http://purl.pt/24995). The choir-books that once belonged to the convent of the Lóios in Évora have been lost or await identification. Other musical manuscripts are a three-volume sanctorale written in 1718-1720 (P-Ln, L.C. 6-8) and two miscellaneous volumes, Figure 1 based on an original 1727 layer, largely modified and amplified up to 1820 (P-Ln L.C.11-12). Selected musical excerpts Orational-hymnary of the Lóios at Braga, Seminário Conciliar (P-BRsc) can be found in the Directorium ad omnes festivitates, in quibus officiari debent praesules hujus coenobii Sanctae Crucis The only folio left from the original fifteenth-century layer Lamecensis, from 1782 (P-Ln Enc. 87). Finally, there is a Manuale chori printed in 1801 (see table 2). (Photo by the author)

444 445 CANTUS PLANUS 2014 FERREIRA — VENETIAN INFLUENCE IN FIFTEENTH-CENTURY PORTUGAL

of polyphony, both written (cantus planus binatim) and improvised. The Portuguese [1] DE IUDICIO MUSICE. C[APITULUM] .IIIJ. canon said to have adopted the liturgy and ritual of San Giorgio, Joham d’Arruda, was an experienced musician. You may have noticed that the only Italian source containing /25 [2] Non est iuditium mu / sice solum in sono, quia / senssus 2 sensus corporis ed. crit. | nam si S | auditus potest falli, ut et / ceteri corporis senssus; nam fa / conspicitur ] respicitur BPRS the traditional, recitational melody of the planctus, related to the Portuguese chanting llitur senssus visus, nam si rectum /30 lignum in aqua ponatur, tone, is a Ritual from San Giorgio in Alga. quia res / picitur occ[u]lis tortuosum. [3] Fa[l]li / tur etiam 3 sensus auditus sepius ed. crit. | a sepius senssus auditus / cum delectatur in sono ab omni / proportione W, proparatione S 35 There is a further demonstrable connection between Lóios and the Veneto, this time racione et proportione remoto. / [4] Ob hoc dicebat Boetius: 4 et formas om. S | que sit ] quid sit Sicut / non su[f]ficit in visu conspici colo /res et formas nisi S | sic nec S | nisi etiam ed. crit. | regarding music theory. A book originating within the congregation, now kept at the ectiam investi // [col. b] getur quid sit eorum proprietas, distincte S Braga Seminary, is a palimpsest: the present contents are those of an orational and / sic nec non su[f]ficit tantum / cantillenis delectari nisi 5 hymnary, but one page survives from the earlier, mid-fifteenth century rendition et / investigetur qualiter proportiones / vocum inter sse disiuncte sint. / [5] Iudicium et proprietas musice, / est racio 5 Iudiciam igitur ed. crit. Igitur om. (figure 1, p. 445). This is a fragment from the Lucidarium by Marchetto da Padova, numerorum, que in ipsa cuncta / dispenssat, et quasi domina BPRS | proprietates W | Remigius the second of only two sources probably written by non-Italians. Among the surviving imperat; / nam Remigius: Veritas mu /10 sice est in numeris dicit VW manuscripts of the treatise two were copied in Venice, one in 1465, the other in 1509. proportionum.

Although the variants in the Braga fragment (Table 3) do not easily fit the proposed [1] QUID SIT MUSICA / C[APITULUM] QUINTUM. before 2 interpolation Musica est transmission stemma (Herlinger 1985, pp. 62-63), they sometimes relate to an scientia recte canendi […] BPR 15 extravagant fifteenth-century source of unknown provenance now in Siena, but not to / [2] Musica est ars spec / tabilis et suavis, cuius / sonus S S 17 in celo et in terra modul[a]tur. / [3] Item musica est scientia 3 et proportionibus | mensis | those copied in Venice itself. que in numeris, propor / cionibus, quantitatibus, menssuris, / The word order in the sources is coniu[n]ccionibus, et consonantiis consis / tit. different from Braga Table 3. Transcription of the Braga fragment, with significant variants taken from [4] Guido: Musica est motus /20 vocum per arssim et thesim, Herlinger’s critical edition hoc / est per elevacionem et deposicionem.

[1] UNDE IUDICATUR / MUSICA. C[APITULUM] VI. 1 Unde dicatur Musica all < [1] DE UTILITATE MUSICE. CAPITULUM TERCIUM. manuscripts / [2] Musica dicitur ad moys, /25 quod est aqua, eo quod 2 before Remigius ut om. MBrCgDLVW [2] Isidorus: Sine musica nulla disciplina potest esse perfecta; iuxta aquas / inventa fuerit, ut Remigius / refert; quia sicut | quod non moveatur […] quod non nichil enim sine illa. [3] Legitur quod David citharam cum aqua non potest tangi / quin moveatur, sic nec musica / quin audiatur BPR percutiebat, Saul minus vexabatur a spiritu, et hoc erat ob audiatur. [3] Inquid enim ipse: /30 Moys genus est quod aquam 3 signat ed. crit. significat dulcedinem musicalis soni. [4] Legitur etiam quod Helyseus 4 habere ] esse W | et before iussit sig[ni] / ficat. [4] Multum enim iuvat aqua / sonum must probabiliter ap[p]a / implentur aqua ut sonum reddant.> [col. a] bere spiritum prophetie, et iussit sibi ps / alterium adduci; ARTE [D]E CANTO et cum psal / lere [in]cepisset, cognoxit spiritum proph / etie in eum mox descendi[s]se, /5 et prophetavit. [5] Videmus namque / cotidie de utilitate musice, / [6] nam preliantes robustiores / fiunt in preliis audientes so / num tubarum et aliorum instrumen /10 torum; similiterque a[n]imalia letantur Venice, however, is made up of more than just its mainland, and includes a small island / ob ipsum sonum. [7] Medici quoque / motum pulssus 7 musicam diiudicant R once populated by ascetic canons. We can suppose that the monastery’s library at San rarum et denssum / esse per ipsam musicam diiudi / cant. [8] 8 rithimos armonie ed. crit. | Giorgio in Alga was initially modest; but it was expanded in 1446 through an important Unde dicebat Pictago /15 ras: Hunc mundum per musicam / rithimicos MBrCgDFLVW | musicas donation by one of the founders, Cardinal Antonio Correr, which encompassed more condidctum esse et gubernari posse, / quia quicquid loquimur rytimas S vel intrinss / ecus venarum pulssibus commo / vetur, per than 120 manuscripts transferred from Padua. The suppression of the congregation musicos rithimi /20 cos armonie virtutibus pro / batur a[s] in 1668, with the corresponding dispersal, and the fire that consumed the convent sociatum esse. [9] Boetius: / Nulla enim magis ad animum / de in 1716 make it difficult to trace its books.18 But when the former chapel-master di[s]ciplinis via quam auribus patet. Joham d’Arruda visited San Giorgio in 1452-1453, he may have found the treatise by

17. The edition lists fifteen complete (or almost complete) manuscripts and three truncated ones. The sources referred to in the text are: of unknown provenance, Siena, I-Sc L.V. 30, fols.. 91v-119 [S]; and two certainly copied in Venice – Perugia, 18. Granata 2010, pp. 193-194. Some of the documents concerning the congregation of San Giorgio in Alga were contained I-PEc 1013 (M.36) [Pg] and Washington, US-Wc Music Division, ML 171.J6 [W]. For the remaining sigla, see Herlinger’s in the Cancelleria della Nunziatura Veneta from 1668, and since 1837 they have been held at the Vatican, as part of its book. Archivio Secreto: cf. Bassetti (2007).

446 447 CANTUS PLANUS 2014 FERREIRA — VENETIAN INFLUENCE IN FIFTEENTH-CENTURY PORTUGAL

Marchetto in the recently expanded library, and decided to copy it for his brethren. – (1998), Leal Conselheiro. Edição crítica, introdução e notas de Maria H. L. de Castro, Lisboa: Imprensa The palaeographical features are compatible with a date in the mid-fifteenth century, Nacional. and both occasion and agent converge to explain the presence of such a treatise among the Portuguese canons. Ferreira Manuel Pedro (2006), Notation and Psalmody: a Southwestern Connection? in Cantus planus. Papers Read at the 12th Meeting of the Ims Study Group, (Lillaführed/Hungary 23-28 August 2004) Budapest: Hungarian Academy of Sciences, pp. 621-639. It is possible that more will be discovered about the Venetian connection to Portugal. ­– (2008), Antologia de Música em Portugal na Idade Média e no Renascimento, 2 vols., 2 CDs, Lisboa: In the last years of the fifteenth century, a cult sprang up in Oporto around some Arte das Musas - Cesem. relics of Saint Pantaleon, allegedly coming from Armenia (Rocha 1970-1971; Mendes 2003). I wonder if Venice, with its church of Saint Pantaleon and its community of ­– (2011) Medieval Music in Portugal Within its Interdisciplinary Context (1940-2010) in José Mattoso ed., The Historiography of Medieval Portugal (c. 1950-2010) Lisboa: Instituto de Estudos Medievais, Armenian merchants, inspired the timely invention of a patron for Oporto, also a city pp. 111-129. of merchants and navigators then in search of recognition. Pursuing the question is, – (2014), Observações sobre o regimento e o enquadramento horário da Capela de D. Duarte in Catarina however, beyond the scope of the present paper. F. Barreira - Miguel M. de Seixas eds., D. Duarte e a sua época: arte, cultura, poder e espiritualidade, Lisboa: Ponte Romana Edições-Iem-Clegh, pp. 29-47.

– (2015), Solange Corbin et les sources musicales du Portugal in Christelle Cazaux-Kowalski - Jean Gribenski - Isabelle His eds., Solange Corbin et les débuts de la musicologie médiévale, Rennes: Presses Universitaires de Rennes, pp. 77-88. Bibliography Fonseca Modesto Flávio Chagas (2003), ‘Motetos para Semana Santa: Subsídios para catalogação’, Cadernos Álvares João (1730), Chronica dos feytos, vida, e morte do Infante santo D. Fernando, que morreo em Fez, do Colóquio 6/1, pp. 50-59. Lisboa Occidental: Miguel Rodrigues. Gallicciolli Giovanni Battista (1795), Delle Memorie Venete Antiche Profane ed Ecclesiastiche­ […] Libri Tre, Andrade Lucas de (1653), Manual das Ceremonias do Officio Solemne da Semana Sancta, Lisboa: Antonio Tomo IV, Venezia: Domenico Fracasso. Alvarez. Granata Giovanna (2010), I libri dei canonici secolari di S. Giorgio in Alga nella documentazione della Azevedo José Manuel Semedo (1960), Procissões da Semana Santa e de Domingo de Páscoa não contidas no Congregazione dell’Indice in Edoardo Barbieri - Federico Gallo eds., Claustrum et armarium. Missal Romano: Guia litúrgico segundo costumes centenários de Portugal, Faro: n.pr. Studi su alcune biblioteche eccesiastiche italiane tra Medioevo ed Età moderna, Milano: Biblioteca Ambrosiana - Roma: Bulzoni, pp. 185-254. Bassetti Massimiliano (2007), Anagrafe di notai veronesi (ASV, Fondo Veneto, 6.724-7.957) in Andrea Castagnetti - Antonio Ciaralli - Gian Maria Varanini eds., Medioevo. Studi e documenti, 2 vols., Haggh Barbara (2009), Ciconia’s Nova Musica: a Work for Singers in Renaissance Padua in João P. Verona: Libreria Universitaria Editrice, 2, pp. 263-280. d’Alvarenga - Manuel P. Ferreira eds., «New Music», 1400-1600: Papers from an International Colloquium on the Theory, Authorship and Transmission of Music in the Age of the Renaissance Calado Adelino Almeida (1993), João Álvares, Frei in Giuseppe Tavani - Giulia Lanciani eds., Dicionário da (Lisbon-Évora, 27-29 May 2003), Évora: Casa do Sul-Centro de História da Arte e Investigaçâo Literatura Medieval Galega e Portuguesa, Lisboa: Caminho, p. 332. Artística da Universidade de Évora - Lisboa: Centro de Estudos de Sociologia e Estética Musical da Universidade Nova de Lisboa, pp. 7-24. Castagna Paulo (2001), A procissão do enterro: uma cerimônia pré-tridentina na América Portuguesa in Istvan Jancso - Iris Kantor eds., Festa: cultura e sociabilidade na América Portuguesa, 2 vols., São Herlinger Jan W. (1985), The Lucidarium of Marchetto of Padua: A Critical Edition, Translation, and Paulo: Imprensa Oficial-Hucitec-Edusp-Fapesp, 2, pp. 827-856. Commentary, Chicago-London: The University of Chicago Press.

Corbin Solange (1947), ‘L’office portugais de la Sepultura Christi’, Revue de musicologie 29, pp. 63-71. Matos José Sarmento de (2009), A invenção de Lisboa, Lisboa: Temas e Debates, vol. 2: As vésperas.

– (1952), Essai sur la musique religieuse portugaise au Moyen Age (1100-1385), Paris: Les Belles Lettres. Mendes Nuno Resende (2003), «Corpo em estilhaços»: o culto patronal a São Pantaleão «do Porto» in Ana P. Machado ed., Esta é a cabeça de São Pantaleão, Lisboa: Ipm, pp. 124-182. ­– (1960), La déposition liturgique du Christ au Vendredi Saint – Sa place dans l’histoire des rites et du théâtre religieux (Analyse de documents portugais), Paris: Les Belles Lettres - Lisbonne: Bertrand. – (n.d.), ‘São Pantaleão «do Porto»: Um paradigma de invenção de relíquias em finais da Idade Média’, http://www.academia.edu/225388/. Cornaro Flaminio (1758), Notizie storiche delle chiese e monasteri di Venezia e di Torcello, Padova: Manfrè. Morais Cristóvão Alão de (1943-1948), Pedatura Lusitana (nobiliário de famílias de Portugal), 12 vols., Cracco Giorgio (1959), ‘La fondazione dei canonici secolari di San Giorgio in Alga’, Rivista di storia della Porto: Livraria Fernando Machado. chiesa in Italia 13, pp. 70-81. Moro Lucia - Cattin Giulio (1993), Il codice 359 del Seminario di Padova (anno 1505): Canti liturgici a due Cunha Dom Rodrigo da (1634-1635), História eclesiástica dos arcebispos de Braga, Braga: Cardoso. voci e laude dei canonici di San Giorgio in Alga in Giulio Cattin - Antonio Lovato eds., Contributi per la storia della musica sacra a Padova, Padova: Istituto per la Storia Ecclesiastica Padovana, pp. Duarte Dom (1982), Livro dos Conselhos. Edição diplomática por João J. A. Dias, Lisboa: Editorial Estampa. 141-189.

448 449 CANTUS PLANUS 2014

Orsolya Csomó Nádas John (2007), The Internationalization of the Italian Papal Chapels in the Early Quattrocento in Franco Piperno - Gabrella Biagi Ravenni - Andrea Chegai eds., Cappelle Musicali fra Corte, Stato e Chiesa nell’Italia del Rinascimento, Firenze: Olschki, pp. 247-269. The feasts of the Holy Virgin in the liturgy of the mass

Palazzo Éric (2015), Le dialogue entre l’histoire de la liturgie et la musicologie: l’apport de Solange Corbin in Zagreb Cathedral and its mediterranean context in Christelle Cazaux-Kowalski - Jean Gribenski - Isabelle His eds., Solange Corbin et les débuts de la musicologie médiévale, Rennes: Presses Universitaires de Rennes, pp. 63-74.

Pina Isabel Pessoa Castro (2011), Os Lóios em Portugal: Origens e primórdios da Congregação dos Cónegos Seculares de São João Evangelista, Ph. D. dissertation, Universidade Nova de Lisboa. The specific, conservativenature of the mass liturgy at the cathedral in Zagreb and its Portalegre Paulo de (2007), Novo Memorial do Estado Apostólico (Primeira Crónica dos Lóios), ed. Cristina survival in the wake of the Council of Trent have been the subject of frequent scholarly Sobral, Lisboa: Roma Editora. research. This paper looks beyond the remarkable uniformity of the Zagrebian rite Reis, António Thomaz dos (1837), Methodo da Liturgia Bracharense, Braga: Typographia Bracharense. to focus on the many different Marian feast days celebrated at the Zagreb cathedral

Ribeiro João Pedro (1810), Dissertações chronologicas e criticas, Lisboa: Academia Real das Sciencias. and to investigate factors that may have influenced its liturgy. A particularly rich set of primary sources from the fourteenth to the eighteenth centuries is available Rocha Pedro Romano (1970-1971), ‘Um breviário bracarense na Biblioteca do Escorial’, Lusitania Sacra 9, to scholars.1 Although seemingly straightforward, selecting the appropriate sources pp. 41-54. for analysis is actually quite complex, for many parts of the mass ordinary are also Santa Maria Francisco de (1697), O Ceo aberto na Terra. Historia das sagradas congregações dos Conegos contained in the Marian masses (e.g., Vultum tuum appears not only in the Feast of Seculares de S. Jorge em Alga de Venesa, & de S. João Evangelista em Portugal, Lisboa: Manoel Lopes Ferreyra. the Circumcision, but also in the masses of the Virgin that fall between it and the Feast of the Purification). I have therefore chosen to deal exclusively with those feasts Santo Tomás Leão de (1974), Benedictina Lusitana, [Coimbra, 1651], ed. José Mattoso, 2 vols., Lisboa: listed in the Marian liturgical calendar. Then, because it has already been subjected Imprensa Nacional-Casa da Moeda. to thorough treatment in previous studies, I omit the Feast of the Purification from my Santos João Duarte dos (1734), Thezouro de ceremonias, que contem as das missas rezadas, e solemnes, assim examination and concentrate on the remaining extant Marian feasts in the liturgy of das festas, como de Defuntos, Braga: F. Duarte da Matta. the mass in Zagreb, arranged in order of the liturgical calendar as follows: Soveral Manuel Abranches de (2004-2014), Ascendências Visienses: Ensaio genealógico sobre a nobreza de Viseu. Séculos XIV a XVII, 2 vols., com Errata e aditamentos, Porto: MASmedia, (http://www.soveral. December 8, Immaculata Conceptio, referred to in Zagreb as the Feast in Sanctifica­ info/AVerrata.htm). tione Mariae and whose chants are identical to those in Nativitatis Mariae Tavares Pedro Vilas Boas (2001), Lóios in Carlos M. Azevedo ed., Dicionário de História Religiosa de Portugal, Lisboa: Círculo de Leitores, pp. 149-157. July 2, Visitatio, which appears only in MR73 (fol. 138) and d23 (fol. 191v) from the fourteenth century; MR170 (fol. 179), MR26 (fol. 159v) and M23 (fol. 236) from the Tomasini Giacomo Filippo (1642), Annales Canonicorum Secularium S. Georgii in Alga, Udine: Nicolai fifteenth century; in the sixteenth-century printed missal M28596 (fol. 173v); and in Schiratti. the eighteenth-century MR6 gradual (p. 284) Wilson Blake (2009), «Transferring Tunes and Adjusting Lines»: Leonardo Giustinian and the Giustiniana in Quattrocento Florence in Mary J. Bloxam - Gioia Filocamo - Leofranc Holdford-Strevens eds., «Uno August 5, Maria de Nive, appearing only in the printed missal M 28596 fol. 180v gentile et subtile ingenio»: Studies in Renaissance Music in Honour of Bonnie J. Blackburn , Turnhout: August 14, Vigilia Assumptionis Brepols, pp. 547-567. August 15, Assumptio September 8, Nativitas Mariae November 21, Praesentatio Mariae: printed missal M 28596 (fol. 201); gradual MR6 (p. 319)

The project is supported by the Hungarian National Research Fund Nr. 84216.

1. HR-Zaa III.d.182 (henceforth = d182); HR-Zmk MR 6; (= MR6) HR-Zmk MR 13 (= MR13) ; HR-Zmk MR 73 (= MR73); HR-Zmk MR 133 (= MR113); HR-Zaa IV.C.59 (= C59); HR-Zaa III.d.23 (= d23); HR-Zmk MR 26 (= MR26); HR-Zmk MR 168 (= MR168); HR-Zmk MR 170 (= MR170); A-Su M.III.23 (= M23); HR-Zmk MR 62 (= MR62); HR-Zmk M 28596 (= M 28596).

450 451 CANTUS PLANUS 2014 CSOMÓ — THE FEASTS OF THE HOLY VIRGIN IN THE LITURGY OF THE MASS IN ZAGREB CATHEDRAL

Common Marian votive masses for different periods of the liturgical year: d23 (fol. Salve sancta parens 1 only between Purification and Maundy Thursday or between 245v); MR13 (fol. 247); MR133 (fol. 197r-v); MR 26 (fol. 200); M23 (fol. 303); MR168 the Ascension and the first Sunday of Advent (fol. 217v); MR170 (fol. 163); printed missal (fol. 227); d182 (fol. 54); MR6, (p. 371) Salve sancta parens 2 only between Easter and Ascension Marian sequences: some sequences in MR13 (fols. 266, 266v, 268); d23 (fols. 264, 267, 269v, 270); M23 (fols. 330v, 334v-335v, 339r-v); MR26 (fols. 217v, 219v-220, 222); The content of Rorate caeli and Vultum tuum is identical in all of the sources mentioned: richer material in the graduals d182 (fols. 72, 82-83, 85, 90v); MR6 (pp. 455, 462, 534, introit Rorate caeli, gradual Tollite portas, alleluia Prophete sancti, offertory Ave Maria 563, 586, 594-606); and in three missals there is a special list of all sequences pairs gratia plena, communion Ecce virgo concipiet, introit Vultum tuum, gradual Diffusa est according to the common Marian votive masses: C.59 (fol. 114); MR170 (fol. 185); gratia, alleluia Post partum virgo, tractus Gaude Maria virgo, offertory Beata es virgo, and printed missal M28596 (fol. 270v) communion Beata viscera. For the Salve sancta parens mass, two main series can be identified in the Zagrebian sources. They are referred to here as Salve sancta parens 1 Sequences and Salve sancta parens 2. Sequences appear in only half of the sources, and an account of their usage lies beyond the scope of this work. However, several interesting observations do bear Table 1. Salve sancta parens 1 mentioning here: the use of sequences cannot be strictly tied to any particular period Introit Salve sancta parens of time, as they are found in one fourteenth century manuscript (MR13), six fifteenth century ones (d23, d182, C.59, MR26, M23, MR170) and in the eighteenth century Gradual Benedicta et venerabilis gradual (MR6). Also, most of these sources tend to draw on the same seven or eight Alleluia Virga Yesse sequences reserved for Marian feasts, with MR170 containing as many as eleven. Moreover, MR170, C.59, and the printed missal also feature an additional section, Tractus Gaude Maria where alleluia-sequence pairs change with the Church calendar (Advent, Lent, Easter, Alleluia for Ascension time Ascendens Christus in altum after Pentecost, etc.), or according to the different types of feasts (dominicalis, ferialis, sollemnis, simplicis). Another particular characteristic of these rubrics is that there is Alleluia for Pentecost Emitte Spiritum no common ferial mass, and after Pentecost, each feria has its own chants. Apart from Offertory Felix namque occasional lacunas, this series is similar in all three sources. Although many of the Offertory Recordare virgo (optional) chants employed in the three series are familiar to us from elsewhere, their structure and rubrics are quite unique. Communion Ave Regina caelorum mater regis

Rubrics Common Marian votive masses are rubricated in careful detail. In contrast with The only difference occurs in the MR13 and in the printed missal, which provides an sources from Esztergom-Strigonium, which normally contain a single votive mass for optional alleluia Post partum virgo. Salve sancta parens 2 is much more intriguing, the Holy Virgin, different sets of chants are used in Zagreb during different periods of with several differences emerging among the sources. The basic structure can be the liturgical year. Although these chants do not appear in all of the Zagrebian sources, found in MR26, M23 and d23. they are present in most, including: two fourteenth-century sources (MR13, MR133); five fifteenth-century ones (d23, M23, MR26, MR168, MR170); the sixteenth-century Table 2. Salve sancta parens 2 printed missal (M 28596); and the eighteenth-century gradual (MR6). The fifteenth- century gradual (d182) contains a lacuna in this section, but the surviving chants show Introit Salve sancta parens a slight difference between the missals without notation and the musical graduals. The Alleluia Angelus Domini descendit rubrics of the textual sources are very consistent: Alleluia O Maria mater pauperum

Rorate Caeli for Advent Offertory Angelus Domini

Vultum tuum between the octave of Christmas and Purification Communion Regina mundi

452 453 CANTUS PLANUS 2014 CSOMÓ — THE FEASTS OF THE HOLY VIRGIN IN THE LITURGY OF THE MASS IN ZAGREB CATHEDRAL

The introit and the communion are a fixed and common element in all of the sources, Dulcis mater dulci nato with no variations. The other sections in Salve sancta parens 2 draw our attention. Firstly, MR13, MR168 and MR170 include an optional Benedicta et venerabilis gradual Subveni mater pia for the alleluias. At a certain point, the offertory Angelus Domini descendit was extended Sancta Maria through the addition of Felix namque (MR13). Felix namque could also be used in place Imperatrix egregie of the other offertory (MR170), and, in MR 133, it even served as the main offertory chant, with the addition of Recordare virgo on Saturdays. Even more interesting are the alleluia chants. In the basic alleluia series, the double verset Angelus Domini is followed This series is followed by two offertories (already familiar to us from Salve sancta by O Maria mater pauperum. A complicating factor is that the alleluia Virga Yesse shows parens 1: Recordare virgo, containing the tropes Ab hac familia and Felix namque). up repeatedly: in MR13 and MR170 it appears as an additional element, whereas in Next come the three familiar communions for Salve sancta parens masses: Ave Regina MR168 it takes the place of the first alleluia, with Angelus Domini coming second and Caelorum, Regina mundi and Beata viscera. This, then, represents the ‘new style’ of O Maria mater pauperum being omitted altogether. To round out the picture of the gradual. But what are its origins? Folio 54r-v of the fourteenth-century gradual d182 Salve sancta parens masses, two more observations are in order. Vigilia Assumptionis is contains the notated chants listed in table 4. used in all sources, with the only exception being caused by a lacuna in MR170, and its chants are invariable: introit Salve sancta parens, offertory Beata es virgo, communion Table 4. Common Marian masses in d182 Beata viscera, gradual Benedicta et venerabilis, offertory Beata es virgo, communion Beata viscera. Maria de Nive is mentioned only in the printed missal, and its chant series Alleluia Emitte spiritum tuum is nearly identical to that of Salve sancta parens 1, except for the communion, which Offertory Felix namque es – sollemni is taken from Salve sancta parens 2: introit Salve sancta parens, gradual Benedicta et venerabilis, alleluia Virga Jesse, offertory Felix namque, communion Regina mundi. Offertory Recordare virgo – sabbatino Offertory Felix namque es - feriali (different melody) Graduals The common Marian masses are treated very differently in the graduals. Unfortunately, Communion Ave regina caelorum – sollemni the fourteenth-century gradual d.182 has just a single folio with this content, but it Communion Regina mundi – sabbatino allows us to conclude that the way of collecting chant in MR6 is not a new phenomenon. Communion Beata viscera – As to Rorate caeli, the main structure of MR6 is similar to that of the missals, but in feriali keeping with the genre of book, it provides many different alleluias as well as tropes. In contrast, the Vultum tuum mass is mentioned only in rubrics with an introit and a tract (the complete mass is notated for the Circumcision). Contrary to tables 1 and 2 – This series makes it apparent that mentioning different chants in the same place is not here Salve sancta parens does not appear in two different series of proprium chants for a new phenomenon, but while in the fourteenth century they are precisely rubricated, specific liturgical periods, but in a complex series, from which chants could be chosen in the eighteenth-century source the chants are no longer rubricated, but merely for the different parts of the mass. One introit and one gradual are given, but a long collected. The choice is left up to the cantor or the singer. series of alleluias (see table 3) is added to the list. Comparison of the sources from Zagreb and Esztergom Table 3. Alleluias in the common Marian masses in MR6 Having carefully examined the Marian feasts in Zagreb, it is now possible to investigate which of the feasts and chants are comparable with other liturgical traditions. The Regina caeli following have been chosen for evaluation: Visitatio, Maria de Nive, Praesentatio O Maria mater pauperum Mariae, and the Salve sancta parens masses, where, the main Zagrebian form has been

Salve dulcis o Maria used for the purpose of this comparison. To begin with, one should note that, as in the Zagrebian sources, the graduals and missals from Esztergom (Strigonium in Latin) O consolatrix pauperum treat the chants differently. More often than not, the graduals from Strigonium provide Post partum virgo a series of genres of chants (i.e., an alleluia-series, an offertory-series, or a communion-

454 455 CANTUS PLANUS 2014 CSOMÓ — THE FEASTS OF THE HOLY VIRGIN IN THE LITURGY OF THE MASS IN ZAGREB CATHEDRAL series), while the missals contain more detailed rubrics, with precise descriptions of earliest printed missals. The evidence indicates that the feast of the Visitatio divides the liturgical occasions. Though the feast of the Visitatio is present in only seven of the practically the whole of Europe. Although its observance was widespread, the feast thirteen known Zagrebian sources, it is found in nearly all sources from Strigonium.2 was not kept everywhere, resulting in a highly heterogeneous scenario. The Visitatio A closer look at the chants from this feast reveals something very interesting: although appears in Hungarian sources as well as in sources from the areas examined in Spain, the content of the Zagrebian Visitatio is completely unified, it differs from that in most Northern Italy (Aosta, and in printed Ambrosian sources), in the Dominican sources, of the Strigonian sources. The only Hungarian tradition resembling the Zagrebian in Cluny, and in the western areas of France (Cambrai and Bourges for example). mass in this case is that of the Pauline order. In fact, the Pauline series of chants for the Despite being celebrated in Northern Italy, the feast was not kept in Rome until Visitatio is identical to that from Zagreb. after the revised missal was promulgated by Pope Paul v. In Hungary, the Visitatio is present in almost all of the sources, including those of the Paulines. It also appears Table 5. Visitatio in Zagreb, although in only about half of the sources, whereas it is entirely missing from a thirteenth-century village missal from the area around Zagreb. This missal is Zagreb Strigonium Pauline a particularly valuable source for the present study, since it originates from before Gaudeamus omnes Gaudeamus omnes Gaudeamus omnes the Zagrebian liturgy began to feel the influence of the Bishop Augustinus Gazotti’s Benedicta et venerabilis Benedicta et venerabilis Benedicta et venerabilis Dominican taste (Kniewald 1959, pp. 6-10). It therefore preserves distinct elements of the influence of Strigonian use on the Zagrebian liturgy. The influence of Esztergom Alleluia In Mariae Alleluia Ave stillans melle Alleluia In Mariae would later wane, so the absence of the Visitatio in this thirteenth-century manuscript Filiae regum Ave Maria Filiae regum is particularly interesting. However, the fact that the feast became widespread in the fourteenth century could be an explanation.5 Even more complex is the case of Beata viscera Diffusa est gratia Beata viscera the chants. The chant series in the Pauline and the Zagrebian sources are the same, meaning that they derive from the oldest Strigonian liturgy. However, in Esztergom itself, the chants changed, probably after Gregorian Romanisation during the twelfth Further observations can be made. Maria de Nive, which is only found in the printed and thirteenth centuries. The History of Liturgy Study Group at the Eötvös Loránd missal in Zagreb, is much more widespread in Esztergom. The relatively new feast University in Budapest recently uncovered a connection between Regensburg and of the Praesentatio Mariae3 appears in the printed Zagrebian missal and in just three Zagreb through similarities in their sacramentaries (Földváry 2017, p. 34) while Strigonian missals.4 As for the Salve sancta parens, whose details in terms of the seeking a potential connection between the Visitatio in Regensburg and in Zagreb. Zagrebian mass are detailed above, connections with the Strigonian series of chants Their finding was exciting, as it showed that the chantseries in Regensburg-Zagreb- are apparent, despite the lack of uniformity in either set of sources. In fact, the main Rome (before the Council of Trent) and in the Pauline sources are nearly the same. difference in the mass sequence lies in the communion, where the Strigonium sources use Regina mundi or Beata viscera in place of Zagreb’s Ave regina Caelorum mater Table 6. Visitatio regis angelorum. Another difference is that the second alleluia used in the Easter series varies in the two groups of sources. Most of the Zagrebian sources use the alleluia Zagreb Strigonium Pauline Rome Regensburg O Maria mater pauperum, whereas in Strigonium they use the Subveni mater. This Gaudeamus omnes Gaudeamus omnes Gaudeamus omnes Gaudeamus omnes Gaudeamus omnes comparison yields two results. The first is that in the case of the sanctorale, neither the Benedicta Benedicta Benedicta Benedicta Benedicta Zagrebian nor the Strigonian sources are completely unified. The second is that the et venerabilis et venerabilis et venerabilis et venerabilis et venerabilis Zagrebian sources correspond in large measure to the Pauline ones. Alleluia In Mariae Ave stillans melle Alleluia In Mariae Alleluia In Mariae Alleluia benignitas benignitas benignitas Spes datur Comparison with the Mediterranean area Alleluia Spes datur As a next step, I extended my research to a greater European, mainly Mediterranean, (optional) context. I selected sixty-five sources to study, half of which were manuscripts, and half Filiae regum Ave Maria Filiae regum Filiae regum Filiae regum

Beata viscera Diffusa est gratia Beata viscera Beata viscera Beata viscera 2. Exceptions: H-Efkö MSS.I.3 (Ulászló gradual); A-GÜ 1/43; RO-AJ R. I. 96.

3. Canonized by Sixtus iv in 1473.

4. H-Bn Inc 176 (printed Pauline missal), H-Bn Clmae 218, H-Bn Clmae 222 (Posonian missals). 5. In Prague it had been canonized in 1385 (Dragoun 1998, p. 157).

456 457 CANTUS PLANUS 2014 CSOMÓ — THE FEASTS OF THE HOLY VIRGIN IN THE LITURGY OF THE MASS IN ZAGREB CATHEDRAL

Maria de Nive is a very important feast in the liturgy of Rome, but it was not widespread For all the diversity of the evidence, some conclusions can be drawn. Marian feasts are in Europe until the Tridentine Reform. Its strong presence in the Roman missals did among the most variable feasts in the liturgical sources, differing even in the extremely not make it popular everywhere. In Hungary, it does occur in the central sources, and unified liturgy of the Zagreb cathedral. Most feasts appear only in some of the sources, mainly the later ones. In the Mediterranean area, Maria de Nive was most common in or in only one of them. Their transmission from other traditions is not clear, but Rome (but not in other Italian centres), and in Spain. As recently pointed out by Zsuzsa examination of quite a number of sources allows us to make some hypotheses. In some Czagány (2013), Aquileia is one of the few areas where its office is well represented; cases, similarity with Pauline sources indicates the provenance of a given feast in the similarly, this feast also appears in three Aquileian graduals: I-Vnm 125, I-CFm 58, and early Strigonian liturgy, although Strigonium may have departed from that tradition I-CFm 79 (Camilot-Oswald 1997). In Hungary, the presence of the Maria de Nive could later on, as with the chants from the Visitatio. In other cases, the Dominican influence be the result of a later Roman influence, although its presence in Pauline sources is evident, as with the Salve sancta parens masses. From the Mediterranean sphere, shows that it was part of the early Strigonian liturgy. In the case of Zagreb, we cannot only the influence of Rome (before the Council of Trent) can be ascertained (see the talk about a real continuity of tradition, as the feast only appears in the printed missal. Salve sancta parens mass). These are the paths of influence that may have resulted in Praesentatio Mariae is a special late feast with sporadic spread. The cases of which the complexity of liturgical practice in Zagreb. It is expected that similar inquiry into we are aware (printed Pauline, two of the Posonian missals, Cordoba, Mallorca, Aosta, the sanctorale may lead to further insights. Paris – but not Dominican),6 do not allow us to draw any conclusions about how it came to be included in the printed Zagrebian missal. We have already covered the use of Salve sancta parens in Hungary. The next interesting step is to identify parallels to it from outside of Hungary and to understand how these came about. It is likely that the liturgy in Zagreb was subject to four spheres of influence: Esztergom, the Dominicans, Regensburg, and the Mediterranean area. Many aspects of these influences converged in a missal called the Missale Spalatense, whose probable provenance was recently identified by Hana Breko-Kustura7 as the Zadar bishopric. The Missale Spalatense was written for use in the area, although it contains Dominican content and includes Hungarian saints as well. It is therefore unsurprising that the series of chants in Salve sancta parens, rubricated for from Purification to Easter, is nearly the same as that in Zagreb, with the sole exception of the Communion. The question is whether this is due to a Dominican, Strigonian or Mediterranean influence. The somewhat surprising answer can be found in the printed Roman missals from the end of the fifteenth century, which have exactly the same order of chants as the Missale Spalatense. On the one hand, this means that Romanisation of the liturgy could have resulted in this series of chants. On the other, it means that the communion is the most variable element in this case, as already seen in the communion variant in Esztergom. The Zagrebian communion Ave regina caelorum mater regis angelorum raises another interesting point. While its use in Hungary is limited to marginal sources (mainly with Bohemian connections), it is well-known in Central Europe and in the Dominican liturgy. So this could be a small point where the Dominicans made an impact on Zagreb. After all of these comparisons, one chant appears to have no parallels: the new-style sixth mode alleluia Maria mater pauperum. This alleluia is found only in the Zagrebian liturgy; it is not even contained in the Monumenta Monodica Medii Aevi.

6. H-Bn Inc 176 (printed Pauline missal), H-Bn Clmae 218, H-Bn Clmae 222 (Posonian missals), Cordoba, Simon Carpintero - Alexus Cardeña, 1561 (Madrid, Biblioteca Nacional de España); Venezia, Lucas Antonius de Giunta, 1506 (Mallorca); see Amiet (1991).

7. See Hana Breko-Kustura (2013).

458 459 CANTUS PLANUS 2014 CSOMÓ — THE FEASTS OF THE HOLY VIRGIN IN THE LITURGY OF THE MASS IN ZAGREB CATHEDRAL

Manuscripts Bibliography

Amiet Robert (1991) Rituale Augustanum, 2 vols., Aoste: Musumeci (Monumenta Liturgica Ecclesiae HR-Zaa III.d.182 gradual fifteenth century Augustanae 12-13). (Zagreb, Arhiv Hrvatske Akademije Znanosti i Umjetnosti) Breko-Kostura Hana (2013), ‘Toward a Possible Origin of the «Missale Romano-Spalatense» Budapest, HR-Zmk MR 6 gradual eigtheenth century National Széchenyi Library, Codex CLMAE 334’, De musica disserenda 9/1-2, pp. 83-97. (Zagreb Metropolitanska Knjižnica) Camilot-Oswald Raffaelle (1997), Die Liturgischen Musikhandschriften aus dem mittelalterlichen Patriarchat HR-Zmk MR 13 missal fourteenth/fifteenth century Aquileia, Kassel-Basel-London-New York-Prague: Bärenreiter (Mmmae, Subsidia 2). (Zagreb Metropolitanska Knjižnica) Czagáni Zsuzsa (2013), ‘Mitteleuropäische Offizien zum Fest Beatae Mariae Virginis de Nive’, De musica HR-Zmk MR 73 missal fifteenth century disserenda 9/1-2, pp. 223–240. (Zagreb Metropolitanska Knjižnica) Dragoun Michal (1998), Neznámé články synodálníh statut pražské arcidiecéze? in Ivan Hlavaček - Jan HR-Zmk MR 133 missal fourteenth century Hrdina eds., Facta probant homines. Sborník příspěvků kživotnímu jubileu prof. dr. Zdeňky Hledíkové (Zagreb Metropolitanska Knjižnica) [Contributions to the jubilaeum of Prof. Dr. Zděnka Hledíková], Praha: Scriptorium. HR-Zaa IV.C.59 missal fifteenth century Földváry Miklós (2017), Egy úzus születése - A Chartvirgus-pontifikále és a magyarországi liturgia (Zagreb, Arhiv Hrvatske Akademije Znanosti i Umjetnosti) megalkotása a XI. században, Budapest: Elte Btk.Vallástudományi Központ Liturgiatörténeti HR-Zaa III.d.23 missal fifteenth century Kutatócsoport (Műhelytanulmányok 4). (Zagreb, Arhiv Hrvatske Akademije Znanosti i Umjetnosti) Kniewald Carolus (1959), ‘Officium et missa de Conceptione et Nativitate B.M.V. secundum consuetudinem HR-Zmk MR 26 missal ca. fifteenth century veterem Zagrabiensem’, Ephemerides Liturgicae 73, pp. 3-21. (Zagreb, Metropolitanska Knjižnica) HR-Zmk MR 168 missal ca. fifteenth century (Zagreb, Metropolitanska Knjižnica) HR-Zmk MR 170 missal ca. fifteenth century (Zagreb, Metropolitanska Knjižnica), Georg de Topusko A-Su M.III.23 missal ca. fifteenth century (Unversitätsbibliothek, Salzburg) HR-Zmk MR 62 missal ca. fifteenth century (Zagreb, Metropolitanska Knjižnica) HR-Zmk M 28596, printed missal 1511 (Zagreb Metropolitanska Knjižnica)

460 461 CANTUS PLANUS 2014

VIII Chant and liturgy in latin Southern Italy

462 463 Thomas Forrest Kelly - Katarina Livljanić

Ferial office in the Montecassino antiphoner and in Beneventan sources of the eleventh and twelfth Centuries

The antiphoner I-MC 542 (hereafter MC 542), datable to the second quarter of the twelfth century, contains an incomplete pars hiemalis (from the fourth Sunday of Advent to the office Ad Mandatum on Maundy Thursday). Written in the elegant Beneventan script used at Montecassino during the time of the famed abbot Desiderius (1057-1087) and his successor Oderisius (1087-1105), its musical notation is diastematic, with clefs and drypoint lines sometimes colored in red (F).1 It is a monastic antiphoner which, to judge by its very close correspondence to the contemporaneous ordinal of Montecassino (Kelly 2008) represents the practices of the abbey of Montecassino (and, presumably, its dependencies) in the twelfth century. The question we pose in this article arises from the strange fact that the ferial office in MC 542 is a secular office, even though all the rest of the book is monastic. Why does a clearly monastic antiphoner have a secular ferial office? It seems unusual that in the founding monastery of the Benedictine order, the ferial office would be secular. In medieval office books the ferial office (or the office de psalmista) is situated usually after the octave of Epiphany (Dominica prima post octavam Epiphaniae) and contains the office for each day of the week: it is the core of the repertoire and reflects the basic and most often sung repertoire in the weekly organization of the liturgy, with the goal of singing the entire psalter in one week. These chants are sung on the days upon which no feast falls. Although its repertoire of antiphons and responsories is usually contained in a dozen folios and does not seem to be a very important part of a manuscript, as compared with the lengthy offices of the major liturgical feasts of the temporal and sanctoral section, we should bear in mind that it is the repertoire which gives the basic structure of psalmody, and it is the part which is the mostly likely to be known by heart by the monks or canons in a community (Gy 1984, pp. 546-549; Mallet-Thibaut 1997, 3, pp. 718-739). In what follows we assume a general familiarity with the office, and with the differences between secular and monastic cursus.2 All the proper offices in MC 542 are monastic, and the book corresponds very closely in all other details with other sources of the Cassinese office (Kelly 2008). But is clear that the ferial office in MC 542 is secular. And that ferial office is not an adaptation of ferial chants to the monastic distribution of psalms; the psalm distribution in the ferial office of MC 542 is that of the secular ferial office. A larger question is the origin of this arrangement of the office. That is to say, the ordinal, and the antiphoner, appear to be a revision of

1. Facsimile edition, Livljanić 2014; other studies: Loew 1980, p. 89; Kelly 1989, p. 308; Newton 1999, pp. 70-72. 2. Convenient listings of the arrangements of psalms are found in Gy (1984, pp. 548-549) and Harper (1991, pp. 258-259).

465 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER the office of Montecassino as carried out by the abbot Desiderius and codified by his Antiphons successor Oderisius – the earliest copy of the ordinal dates from Oderisius’ time. We The ferial office here is typical and generally consistent, similar to secular offices found have essentially no office books from before Desiderius’s time, and the reports in the elsewhere. It is relatively easy to determine that the psalms in the ferial office in MC Chronicle of Montecassino, stating that Desiderius renewed of all aspects of the abbey 542 are distributed according to the secular use since in the ferial office the antiphon is (Kelly 2008, pp. 54-57), make it very likely that the office as we have it is a result of his almost always drawn from the psalm (or the group of psalms) it accompanies. And so, reforming work. So the question is: if Desiderius imposed a new, or revised, office on even in the many cases where the manuscript does not give the incipit of the psalm or Montecassino, where did it come from? Did he make it up, or did he import it? And psalms to be sung with an antiphon, we can be sure what the psalm was, and we can be is this secular office, so different from the monastic ferial office reported in all the sure that the psalms are distributed through the week according to the secular cursus. ordinals, really a part of Cassinese practice? So it is not a matter of a monastic community adapting the available antiphons to the monastic distribution of psalms; the psalm distribution in the ferial office of MC 542 is Along with the antiphoner MC 542, several other regional sources have been included that of the secular ferial office everywhere. A look at this ferial office itself shows that it in this study: is internally consistent. Generally, everything corresponds to a well-made ferial week. The antiphons match their psalms; mostly they are well-known standard antiphons. The eight manuscript sources of the ordinal of Montecassino (Kelly 2008, pp. 72-73): Only a few places call for notice. Cassinese ordinals Los Angeles, J. Paul Getty Museum, ms. 83.ML.97 Table 1. Antiphons for Sunday Matins in Montecassino 542 Montecassino, Archivio dell’Abbazia, I-MC ms. 198 Montecassino Archivio dell’Abbazia, I-MC ms. 562 Vigilia i Paris, Bibliothèque Mazarine, F-Pm ms. 364 Beatus vir Ps 1 Vatican City, Biblioteca Apostolica Vaticana, V-CVbav ms. Urb. Vat. lat. ms. 585 Servite Ps 2, 3, 6 Beneventan ordinals, adapted from those of Montecassino Benevento, Biblioteca capitolare, I-BV ms. 66 Domine deus Ps 7, 8, 9, 10 Naples, Biblioteca Nazionale, I-Nn ms. VI E 43 Illumina Ps 11, 12, 13, 14 Vatican City, Biblioteca Apostolica Vaticana, V-CVbav Vat. lat. ms. 4928 Vigilia ii I-MC Compactiones V, fragment of a notated monastic breviary, twelfth century Conserva Ps 15 I-BV 19, secular-cursus notated breviary and missal, twelfth century (Mallet-Thibaut Bonorum Ps 15! 1997, 2, p. 65; Kelly 1989, p. 299). Propter verba Ps 16 I-BV 21, monastic antiphonary, twelfth-thirteenth century (Mallet-Thibaut 1997, 2, pp. 71-75; Kelly 1989, p. 300; Hesbert 1963-1979) Diligam Ps 17

Vivit dominus (no Ps indicated) MC 542 corresponds closely to the ordinal, except in the ferial office, where the ordinals uniformly give a monastic office – one that corresponds also to the office in the Vigilia iii fragments found in the Montecassino Compactiones and to the monastic offices in I-BV Non sunt Ps 18 21 and the other monastic sources from the region. The only regional manuscript with a secular cursus, the breviary I-BV 19, corresponds quite closely to the ferial office in Exaudiat Ps 19 MC 542, as we shall see. Domine in virtute Ps 20

Dns tamquam ovis (no Ps indicated) There is much to consider here; we propose to set aside many interesting aspects – hymns, invitatories, readings – and concentrate on the antiphons and the responsories Alleluia (× 10, with of this office, hoping to find a context and a background for it. melodies indicated)

466 467 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER

Sunday matins has a couple of irregularities: the psalm distribution, skipping psalms 4 tua) matches the Beneventan sources, not the Cassinese (which have Neque irrideant). and 5, is standard. But the antiphons are a bit unusual. There are two antiphons drawn The MC 542 responsories in the last nocturn come from psalms 23-25, which are sung from psalm 15 – perhaps they are meant as alternatives? – and two antiphons with no on Sunday only in the monastic cursus. But its order is still the closest to I-BV 19 which, psalmody specified. Vivit dominus may be meant as an alternative for psalm 17, since although it has two of these responsories (nos. 7 and 8) is also secular. On Monday the its text is drawn from verse 47; and possibly the same is intended for Dominus tamquam differences between the sources are more noticeable. I-BV 19 is the closest to MC 542, ovis, although its text is not drawn from psalm 20; there is no need for anything beyond which adds one responsory at the beginning: Adjutor meus esto deus. The responsory psalm 20. The Alleluias are surely intended as alternatives in Eastertide. With these Delectare in domino (Cao 6404), used by the other sources (but absent from MC 542), exceptions, this is a well-made ferial office. But where does it come from? We have comes from psalm 36, sung at Monday matins. For Tuesday several responsories very little to go on. The ferial office here does correspond closely to the only other in MC 542 have verses not identified by Cao. The first responsory, Auribus percipe full source of the secular office in southern Italy, namely I-BV 19, already mentioned. domine, has a rare verse Exaudi orationem meam (found at the same position in the Whether Montecassino borrowed its ferial office from Benevento cathedral may be antiphoner from Marseille Cathedral, F-Pn lat 1090). All the other southern Italian doubted, however, because of the differences, small as they may seem, between the manuscripts have the responsory Tibi soli peccavi (from psalm 50, sung at lauds). two offices. The antiphons of the two offices are compared in table 2, p. 472. So MC 542 is the only one in the group to have all the responsories from the matins psalms of the day. On Wednesday, the list of responsories in MC 542 corresponds to Most cases involve a different choice of well-known antiphons for psalms. Here we will the Tuesday list in the ordinals and I-BV 21. The first responsory in MC 542 is Tibi not comment on variants of the canticle antiphons, which generally are chosen from soli peccavi, from psalm 50 (sung at lauds). The second responsory has a non-psalmic the text of Magnificat and Benedictus, and can be switched at will. A few antiphons are text Ne perdideris me, domine: monastic manuscripts use it mostly on Tuesday, and rare, and point to local sources: the secular manuscripts on Wednesday. Adiutor meus is from psalm 58, which is sung on Wednesday matins in the secular cursus. On Thursday all responsories in MC 542 On Sunday, the antiphon Illumina of MC 542 at matins is very rare, and the vespers come from psalm 70, while the other regional sources include responsories from other antiphon Magna opera domini is limited to MC 542 and to monastic sources from psalms as well (72, 74, 76, 79). Benevento. Monday matins has another rare antiphon: Sedebit dominus rex, for psalm 28 (and On Friday all the sources have an identical list of responsories and verses (from psalms 29), is also known here and at Benevento. 85 and 93). This uniformity is natural on this particular day of the week, since both monastic and secular cursus use these same psalms at matins. The responsory list for Noteworthy at Wednesday matins is not so much the variety of well-known antiphons Saturday is identical between MC 542 and I-BV 19. as the non-psalmic antiphon Deus vita mea, known only from these two manuscripts. For Friday lauds I-BV 19 is the only Italian manuscript to use the relatively rare Conclusions Redde mihi for psalm 50. It seems clear that these two ferial offices, Montecassino and Benevento, are closely related. Neither of these versions of the ferial office is duplicated in any manuscript Despite the few differences, there is clearly some relatively close relationship among known to us, but they are very close to each other. On the basis of the antiphons it the surviving books of southern Italy. would seem that the ferial office of MC 542 draws on an existing local tradition known also at Benevento. In adopting, or borrowing a version of this tradition, the office Responsories of Montecassino makes some changes: it stuffs the Sunday matins with additional The matins responsories in MC 542 and in the southern Italian sources already antiphons, and it substitutes two unique antiphons on Sunday: Illumina and Magna mentioned are shown in table 3, p. 477. Their texts, here and elsewhere are generally opera domine. It is not the same as any surviving source. taken from the psalms which are sung during matins of the day according to secular use (Le Roux 1963, pp. 39-149). In the responsory list, we notice the same similarity between MC 542 and I-BV 19, both witnesses to a local version of a secular cursus. However, there are some interesting On Sunday I-BV 19 uses the series of responsories of MC 542 and other manuscripts, but details: the list of responsories shows more variety on certain days (Tuesday, precedes it with the responsory Super populum, shifting the others one place later. The Wednesday, Thursday), while Friday is a totally uniform day throughout the secular verse of the eighth responsory, Ad te, domine, levavi in MC 542 (Dirige me, in virtute and monastic sources.

468 469 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER

From a paleographical point of view, MC 542 seems to be a product of the Cassinese Bibliography scriptorium. It also seems to be linked to a monastic milieu (Monday and Tuesday Holy Week psalmody in lauds, and the rite of Saturday foot-washing (Livljanic 2014, p. Codex 21 de la Bibliothèque capitulaire de Bénévent: Antiphonale monasticum, XII-XIIIe s. (2001), Solesmes: 34). The Saturday rite of foot-washing is documented in MC 542 in the rubric Sabato Froidfontaine-Abbaye Saint-Pierre, 2001 (PalMus 22). ad mandatum (MC 542, p. 29). This practice is habitual for a monastic use. As further Gy Pierre-Marie (1984), La Bible dans la liturgie au Moyen Âge in Pierre Riché - Guy Lobrichon eds., Le evidence of a secular context, there is a mention of a bishop (instead of the abbot Moyen Age et la Bible, Paris: Beauchesne (Bible de tous les temps 4), pp. 537-552. mentioned in the ordinals) in the Palm Sunday processions (MC 542, p. 178): Harper John (1991), The Forms and Orders of Western Liturgy from the Tenth to the Eighteenth Century, Oxford: Clarendon. Hora tertia omnes fratres, indutis albis lineis eo[dem] veniant in ecclesia et cantent terciam […] Hesbert René-Jean (1963-1979), Corpus antiphonalium officii, 6 vols., Rome: Herder (Rerum ecclesiasti­ Post quos dominus episcopus pluviali[s] indutus, cum ministris, tunicis seu dalmaticis vestitis, carum documenta, series maior, fontes 7-12). procedat et, congregatis omnibus in ecclesia, expositis in medio palmarum atque olivarum ramis, incipiant in choro a[ntiphonam] Osanna filio David[…]. Kelly Thomas F. (1989), The Beneventan Chant, Cambridge: Cambridge University Press.

– (2008), The Ordinal of Montecassino and Benevento: Breviarium Sive Ordo Officiorum, 11th Century, MC 542 is a manuscript that was certainly used over several centuries: several later Fribourg: Academic Press Fribourg (Spicilegium Friburgense 45). scribal indications bear witness to this, e. g. psalm or chapter incipits, additional use of certain antiphons for the little hours or other offices (Livljanic 2014, p. 34). Is it possible Le Roux Raymond (1963), ‘Les répons de psalmis pour les matines de l’Epiphanie à la Septuagésime selon les cursus romain et monastique’, Etudes grégoriennes 6, pp. 39-149. that Montecassino (or the church for which the manuscript was made) intended this secular office to be used on ferial days? That this is an adaptation of the secular office Livljanić Katarina (2014), Montecassino, Archivio dell’Abbazia, ms. 542. Antiphonaire (12ème siècle), Solesmes: La Froidfontaine-Éditions de Solesmes (PalMus 23). for use in the monastery? This seems unlikely, mostly because the rest of the book does not correspond with this distribution of psalms or this arrangement of the office. There Loew Elias A. (19802), The Beneventan Script; A History of the South Italian Minuscule, revised and enlarged are places where the two come into conflict: major feasts all have twelve responsories, edition by Virginia Brown, 2 vols., Rome: Edizioni di storia e letteratura (Sussidi eruditi 33-34). not the nine of the secular office; major feasts have antiphons ad cantica for the third Mallet Jean - Thibaut André (1997), Les Manuscrits en écriture bénéventaine de la Bibliothèque capitulaire nocturn (this does not happen in the secular office); the antiphons of lauds for the first de Bénévent, 3 vols. (1984-1997), Paris: Cnrs - Turnhout: Brepols (cantusdatabase.org/source/374018/ i-bv-19). Sunday in Lent and elsewhere correspond to the monastic distribution of psalms, not the secular. Many such examples occur throughout the antiphoner, which is otherwise Newton Francis (1999), The Scriptorium and Library at Monte Cassino, 1058-1105, Cambridge: Cambridge clearly monastic. So this secular office was imported into a manuscript that otherwise University Press (Cambridge Studies in Palaeography and Codicology 7). reflects very closely the monastic practices of Montecassino. The question remains, then: where did it come from? Perhaps it was somehow composed, at the command of Desiderius. The form of the secular office is of course well known in the Benedictine world, for it is used everywhere for the last three days of Holy Week. It seems unlikely that this adaptation took place at Montecassino, since it seems impossible that this ferial office was really used at the monastery. But the inclusion, either accidental or intentional, of a ferial office in the only surviving antiphoner of Montecassino allows us to see that there was indeed a regional tradition in the secular office; perhaps the future discovery of additional sources will allow us to focus even more closely.

470 471 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER

Table 2 Legenda p. 476

SUNDAY I-MC 542 I-BV 19

I-MC 542 I-BV 19 a. P. Alleluia ii a. P. Gloria tibi trinitas* a. T. Alleluia vii a. T. Adesto deus unus* a. S. Alleluia v (In velamento) a. S. Te semper idem esse Mat. l. Mat. l. a. N. Alleluia iii (Beata) a. N. Te invocamus*

1.1 Beatus vir qui non abiit Vespera Vespera 1.2 Servite domino in timore 1.1 Servite domino in timore 1.3 Domine deus meus in te 1.2 Domine deus meus in te a. Sede a dextris meis dicit a. Sede a dextris meis dicit 1.4 Illumina oculos meos domine 1.3 Respice et exaudi me domine a. Magna opera domini a. Fidelia omnia mandata ejus a. In mandatis ejus cupit nimis a. In mandatis ejus cupit nimis 2.1 [Conserv]a me domine quoniam a. Sit nomen domini benedictum a. Sit nomen domini benedictum 2.2 Bonorum meorum non indiges in 2.1 Bonorum meorum non indiges in a. Nos qui vivimus benedicimus a. Nos qui vivimus benedicimus 2.3 Propter verba labiorum meorum 2.2 Propter verba labiorum meorum 2.4 Diligam te domine virtus mea 2.3 Diligam te domine virtus mea a. M. Dixit Jesus ministris* a. M. Deficiente vino iussit 2.5 Vivit dominus et benedictus MONDAY 3. Non sunt loquelae neque 3.1 Non sunt loquelae neque 3. Exaudiat te dominus in die 3.2 Exaudiat te dominus in die Mat. l. Mat. l. 3. Domine in virtute tua 3.3 Domine in virtute tua 3. Alleluia vi (Dominus tamquam) a. Dominus defensor vitae meae a. Dominus defensor vitae meae 3. Alleluia vi (Oblatus est) a. Sedebit dominus rex in a. Adorate dominum in aula 3. Alleluia ix (Lux de luce) a. In tua justitia libera me a. In tua justitia libera me 3. Alleluia vii (Omnis nationes) a. Rectos decet collaudatio a. Rectos decet collaudatio 3. Alleluia viii (Stella quam) a. Expugna impugnantes me a. Expugna impugnantes me 3. Alleluia v (Michael) a. Revela domino viam tuam [sic] a. Revela domino viam tuam [sic] 3. Alleluia vi (Michael praep-) 3. Alleluia ix (Michael Gabriel) 3. Alleluia vi (Mentem sanctam) Laudes Laudes […]. Alleluia viii (Quidam autem) a. Miserere mei deus a. Miserere mei deus a. Intellege clamorem meum a. Intellege clamorem meum Laudes Laudes a. Deus deus meus ad te de luce a. Deus deus meus ad te de luce a. Conversus est furor tuus a. Conversus est furor tuus a. Regnavit dominus praecinctus a. Regnavit dominus praecinctus a. Laudate dominum de caelis a. Laudate dominum de caelis a. Sciamus omnes quia dominus a. Sciamus omnes quia dominus a. Benedicam te domine in vita a. Benedicam te domine in vita B. Benedictus deus Israel B. Benedictus deus Israel a. [Tres pueri] jussu* a.[Tres pueri] jussu* a. Spiritus omnis laudet dominum a. Spiritus omnis laudet dominum a. P. Beati qui ambulant in lege a. T. Deduc me domine in semitam a. T. Deus exaudi orationem meam ad Benedicite: a. S. Adjuva me et salvus ero a. S. Adjuva me et salvus ero a. Tres pueri jussu regis in a. N. Aspice in me et miserere mei a. N. Aspice in me et miserere mei a. Tres ex uno ore cantabant re- a. Non cessabant ministri regis a. In camino ignis pueri tres Vespera Vespera a. Sidrac Misac Abdenago magni a. Anania Azaria Misael dominum a. Inclinavit dominus aurem suam a. Inclinavit dominus aurem suam a. Laudemus viros gloriosos qui a. Credidi propter quod locutus a. Credidi propter quod locutus a. Video virum similem filio dei a. Laudate dominum omnes gentes a. Laudate dominum omnes gentes a. Hymnum dicamus alleluia a. Clamavi et exaudivit me a. Clamavi et exaudivit me a. Caminus ardebat septies quam a. Unde veniet auxilium mihi a. Auxilium meum a domino a. B. Nuptiae factae sunt in Cana a. M. Magnificet te semper anima a. M. Magnificat anima mea dominum

472 473 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER

TUESDAY I-MC 542 I-BV 19

I-MC 542 I-BV 19 Vespera Vespera

Mat. l. Mat. l. a. Nisi tu domine servaveris nos a. Nisi tu domine servaveris nos a. Beati omnes qui timent a. Beati omnes qui timent a. Ut non delinquam in lingua a. Ut non delinquam in lingua a. Saepe expugnaverunt me a a. Saepe expugnaverunt me a a. Sana domine animam meam quia a. Sana domine animam meam quia a. De profundis clamavi ad te a. De profundis clamavi ad te a. Eructavit cor meum verbum a. Eructavit cor meum verbum a. Speret Israel in domino a. Speret Israel in domino a. Adjutor in tribulationibus a. Adjutor in tribulationibus a. Magnus dominus et laudabilis a. Auribus percipite qui a. M. Respice humilitatem meam a. M. Respice humilitatem meam a. Deus deorum dominus locutus a. Deus deorum dominus locutus

THURSDAY Laudes Laudes Mat. l. Mat. l. a. Dele domine deus iniquitatem a. Dele domine deus iniquitatem a. Salutare vultus mei deus meus a. Salutare vultus mei deus meus a. Quaerite dominum et vivet a. Domine deus in adiutorium a. Ad te de luce vigilo deus a. Ad te de luce vigilo deus a. Esto mihi domine in deum a. Esto mihi domine in deum a. Cunctis diebus vitae nostrae a. Cunctis diebus vitae nostrae a. Ponere in deo spem meam a. Ponere in deo spem meam a. Sol et luna laudate deum a. Sol et luna laudate deum a. In Israel magnum nomen ejus a. In Israel magnum nomen ejus a. Tu es deus qui facis a. Tu es deus qui facis a. B. Erexit dominus nobis cornu a. B. Erexit dominus nobis cornu a. Propitius esto peccatis a. Propitius esto peccatis

Vespera Vespera Laudes Laudes a. In domum laetantes ibimus a. In domum laetantes ibimus a. Averte domine faciem tuam a. Averte domine faciem tuam a. Ad te levavi oculos meos a. Ad te levavi oculos meos a. Domine refugium factus es a. Domine refugium factus es a. Adjutorium nostrum in nomine a. Adjutorium nostrum in nomine a. In matutinis domine meditabor a. In matutinis domine meditabor a. Benefac domine bonis et a. Benefac domine bonis et a. Cantemus domino gloriose a. Cantemus domino gloriose a. Facti sumus sicut consolati a. Facti sumus sicut consolati a. Cantate domino canticum novum a. Cantate domino canticum novum a. M. In deo salutari meo a. M. In deo salutari meo a. B. In sanctitate serviamus a. B. Erexit dominus nobis cornu

WEDNESDAY Vespera Vespera

Mat. l. Mat. l. a. Et omnis mansuetudinis ejus a. Et omnis mansuetudinis ejus a. Ecce quam bonum et quam a. Ecce quam bonum et quam a. Avertet dominus captivitatem a. Intende in me et exaudi me a. Omnia quaecumque voluit a. Omnia quaecumque voluit a. Deus vita mea nuntiavi tibi a. Deus vita mea nuntiavi tibi a. Quoniam in saeculum a. Quoniam in saeculum a. Juste judicate filii hominum a. Juste judicate filii hominum a. Hymnum cantate nobis de a. Hymnum cantate nobis de a. Da nobis domine auxilium de a. Da nobis domine auxilium de a. Nonne deo subdita erit anima a. A timore inimici eripe domine a. M. Deposuit potentes sanctos a. M. Deposuit potentes sanctos a. In ecclesiis benedicite a. Benedicite gentes deum FRIDAY Laudes Laudes Mat. l. Mat. l. a. Tibi soli peccavi domine a. Tibi soli peccavi domine a. Te decet hymnus deus in Sion a. Te decet hymnus deus in Sion a. Exsultate deo adjutori nostro a. Exsultate deo adjutori nostro a. Labia mea laudabunt te in a. Labia mea laudabunt te in a. Tu solus altissimus super a. Tu solus altissimus super a. Dominus judicabit fines a. Dominus judicabit fines a. Benedixisti domine terram a. Benedixisti domine terram a. Caeli caelorum laudate deum a. Caeli caelorum laudate deum a. Intret oratio mea in a. Intret oratio mea in a. Benedictus dominus in a. Benedictus dominus in a. B. De manu omnium qui nos a. B. De manu omnium qui nos a. Cantate domino et benedicite a. Cantate domino et benedicite

474 475 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER

Table 3. Responsories in I-MC 542 and related manuscripts I-MC 542 I-BV 19 Legenda p. 479

Laudes Laudes a. Spiritu principali confirma a. Redde mihi laetitiam SUNDAY - MONDAY a. In veritate tua exaudi me a. In veritate tua exaudi me a. Illumina domine vultum tuum a. Illumina domine vultum tuum I-MC 542 I-BV 19 a. Domine audivi auditum tuum et a. Domine audivi auditum tuum et I.3 R. Super populum tuum domine (7725) a. In sanctis ejus laudate deum a. In sanctis ejus laudate deum SUNDAY I.3za V. Exsurge domine salvum me fac (7725za) a. B. Ad dandam scientiam plebi a. B. Ad dandam scientiam plebi 1.1 R. Domine ne in ira (6501) Ps 6 1.1 R. Domine ne in ira tua arguas 1.1a V. Timor et tremor (6501a) 1.1a V. Timor et tremor venerunt Vespera Vespera 1.2 R. Deus qui sedes super thronum (6433) Ps 9 1.2 R. Deus qui sedes super thronum 1.2a V. Tibi enim derelictus est (6433a) 1.2a V. Tibi enim derelictus est a. In conspectu angelorum a. In conspectu angelorum 1.3 R. A dextris est mihi dominus ne (6002) Ps 15 1.3 R. A dextris est mihi dominus a. Domine probasti me et a. Domine probasti me et 1.3b V. Dominus pars hereditatis meae (6002b) 1.3b V. Dominus pars hereditatis a. A viro iniquo libera me a. A viro iniquo libera me 1.4 R. Notas mihi fecisti domine (7240) Ps 15 1.4 R. Notas mihi fecisti domine a. Domine clamavi ad te exaudi a. Domine clamavi ad te exaudi 1.4b V. Tu es domine qui restitues (7240b) 1.4b V. Tu es domine qui restitues a. Portio mea domine sit in a. Portio mea domine sit in 1.5 R. Diligam te domine virtus mea (6453) Ps 17 1.5 R. Diligam te domine virtus mea 1.5b V. Et liberator meus deus meus (6453b) 1.5b V. Et liberator meus deus meus a. M. Suscepit deus Israel puerum a. M. Suscepit deus Israel puerum 1.6 R. Firmamentum meum et refugium (6736) Ps 17 1.6 R. Firmamentum meum et refugium 1.6a V. Protector meus et cornu (6736a) 1.6a V. Protector meus et cornu 1.7 R. Domini est terra et plenitudo SATURDAY 1.7 R. Domini est terra et plenitudo (6517) Ps 23 1.7a V. Ipse super maria fundavit eam (6517a) 1.7a V. Ipse super maria fundavit eam 1.8 R. Ad te domine levavi animam Mat. l. Mat. l. 1.8 R. Ad te domine levavi animam (6026) Ps 24 1.8c V. Dirige me in veritate tua (6026c) 1.8c V. Dirige me in veritate tua a. Quia mirabilia fecit dominus a. Quia mirabilia fecit dominus 1.9 R. Audiam domine vocem laudis (6144) Ps 25 a. Jubilate deo omnis terra a. Jubilate deo omnis terra 1.9a V. Domine dilexi decorum (6144a) a. Clamor meus ad te veniat deus [antiphon omitted] a. Benedic anima mea dominum a. Benedic anima mea dominum MONDAY a. Visita nos domine in salutari a. Visita nos domine in salutari a. Confitebor domino nimis in a. Confitebor domino nimis in 2.1 R. Adjutor meus esto deus (6037) Ps 26 2.1a V. Neque despicias me deus (6037a) 2.2 R. Quam magna multitudo Laudes Laudes 2.2 R. Quam magna multitudo (7459) Ps 30 2.2a V. Perfecisti eam sperantibus in (7459a) 2.2a V. Perfecisti eam sperantibus a. Benigne fac in bona voluntate a. Sacrificium deo spiritus 2.3 R. Benedicam dominum in (6237) Ps 33 2.3 R. Benedicam dominum in omni a. Bonum est confiteri domino a. Bonum est confiteri domino 2.3a V. In domino laudabitur anima (6237a) 2.3a V. In domino laudabitur anima a. Metuant dominum omnes fines a. Metuant dominum omnes fines II.1 R. Delectare in domino et dabit a. Justus et sanctus deus noster [antiphon omitted] II.1b V. Revela domino viam tuam (6404b) a. In cymbalis benesonantibus a. In cymbalis benesonantibus ad B. In viam pacis et salutis ad B. Sedentes in tenebris et umbra

Vespera Vespera a. Benedictus dominus deus meus a. Benedictus dominus deus meus a. Per singulos dies benedicam a. In aeternum et in saeculum a. In aeternum et in saeculum a. Laudabo deum meum in vita mea a. Laudabo deum meum in vita mea a. Deo nostro jucunda sit a. Deo nostro jucunda sit a. Lauda Jerusalem dominum a. Lauda Jerusalem dominum a. M. Sicut locutus est ad patres a. M. Fratres sit vobis cor unum a. M. Sicut locutus est ad patres

Legenda Differences in bold; pieces in I-MC 542 but not in I-BV 19 in Roman. Signs and abbreviations according Cantus. A Database for Latin Ecclesiastican Chant, http://cantusdatabase.org/description]

476 477 CANTUS PLANUS 2014 FORREST KELLY - LIVLJANIĆ — FERIAL OFFICE IN THE MONTECASSINO ANTIPHONER

TUESDAY - THURSDAY FRIDAY - SATURDAY

I-MC 542 Ordinals, I-BV 21 I-BV 19 I-MC 542 Ordinals, Compactiones V, I-BV 19, I-BV21

TUESDAY FRIDAY

3.1 R. Auribus percipe domine 4.1 R. Tibi soli peccavi 3.1 R. Auribus percipe domine 6.1 R. Confitebor tibi domine deus (6317) Ps 85 Same as I-MC 542 (6154) Ps 38 4.1a V. Amplius 3.1z V. Exaudi orationem meam 6.1a V. Et eripuisti animam meam ex (6317a) 3.1z V. Exaudi orationem meam 4.2 R. Ne perdideris me domine 6.2 R. Misericordia tua domine magna (7161) Ps 85 domine (6154z) 4.2b V. Exaudi deus 3.3 R. Ego dixi domine miserere 6.2b V. In die tribulationis meae (7161b) 3.2 R. Statuit dominus supra 4.3 R. Adiutor meus mei 6.3 R. Factus est mihi dominus in (6716) Ps 93 petram (7698) Ps 39 4.3a V. Eripe me 3.3z V. Ab omnibus iniquitatibus 6.3c V. Quia fortitudo mea et laus (6716c) 3.2z V. Et exaudivit deprecationem meis (7698z) 4.1 R. Tibi soli peccavi et malum 3.3 R. Ego dixi domine miserere 4.1a V. Amplius lava me domine I-MC 542, I-BV19 Ordinals, Compactiones V, I-BV21 (6627) Ps 40 ab 3.3z V. Ab omnibus iniquitatibus SATURDAY (6627z) 7.1 R. Misericordiam et judicium (7162) Ps 100 7.2 R. Domine exaudi 7.1a V. Perambulabam in innocentia (7162a) 7.2c V. Ne avertas WEDNESDAY 7.2 R. Domine exaudi orationem meam (6495) Ps 101 7.3 R. Velociter exaudi 7.2c V. Non avertas faciem tuam a me (6495c) 7.3a V. Dies mei 4.1 R. Tibi soli peccavi et malum IV.1 R. Exaudi deus 4.2 R. Ne perdideris me domine 7.3 R. Velociter exaudi me domine (7820) Ps 101 VII.1 R. Salvos nos fac (7565) Ps 105 (7765) Ps 50 (6685) cum 7.3a V. Dies mei sicut umbra (7820a) VII.1a V. Memento (7565a) 4.1a V. Amplius lava me domine IV.1b V. Quoniam tu deus 4.2b V. Exaudi deus orationem ab (7765a) (6685b) meam 4.2 R. Ne perdideris me domine 5.1 R. Deus in te speravi IV.3 R. Paratum cor meum deus cum (7208) NP 5.1b V. Quoniam tu es (7350) Exaudi deus orationem IV.2 R. Mihi autem IV.3a V. Exsurge gloria mea 4.2b V. Legenda (7208b) Ps 54 (7153) Ps 72 (7350a) The responsories are numbered according to their appearance in I-MC 54: 4.3 R. Adjutor meus tibi psallam IV.2a V. Ut annuntiem IV.1 R. Exaudi deus deprecationem First number = day of the week (1 = Sunday, 2 = Monday, etc.) (6038) Ps 58 (7153a) IV.1b V. Quia tu deus exaudisti Second number = place of a responsory in the Matins (1.1 = the first Sunday responsory) 4.3a V. Eripe me de inimicis meis Number in brackets after the incipit = Cao number (6038a) If the responsory text is psalmic, the psalm number is identified (non-psalmic responsories = NP). Where responsories have more than one verse, the Cao number has a letter appended to indicate which verse is meant.

THURSDAY The responsories in the other sources are numbered according to the following principle: If a responsory appears in I-MC 542, it keeps its identification number, whatever its place is in the other manuscripts. If a responsory is absent from I-MC 542 and appears only in other manuscripts, its day of the week is given in Roman 5.1 R. Deus in te speravi V.1 R. Confitebimur 5.1 R. Deus in te speravi numerals. The other number (signifying the place of the responsory in the Matins) is given in Arabic numerals and starts (6423) Ps 70 (6315) Ps 74 (6423) Ps 70 with 1 at the first appearance of that responsory. 5.1b V. Quoniam tu es patientia V.1a V. Narrabo Omnia 5.1b V. Quoniam tu es patientia mea (6423b) (6315a) (6423b) 5.2 R. Repleatur os meum laude V.2 R. Cogitavi dies IV.2 R. Mihi autem tua (7529) Ps 70 (6300) Ps 76 IV.2a V. Ut annuntiem 5.2z V. Deus ne elonges a me deus V.2b V. In die tribulationis 5.3 R. Gaudebunt labia mea (7529z) (6300b) (6762) Ps 70 5.3 R. Gaudebunt labia mea dum V.3 R. Divastavit 5.3a V. Sed et lingua mea (6762) Ps 70 (6435) Ps 79 (6762a) 5.3a V. Sed et lingua mea V.3a V. Domine deus virtutum meditabitur (6762a) (6435a)

478 479 Luisa Nardini

The masses for the Dead in Beneventan manuscripts Issues of formulary organization and chant manipulation

The tradition for the mass of the Dead shows a high degree of variability for what concerns the size, choice, and order of chants and prayers (Gay 1957; Hesbert 1952; Sicard 1978; Valle 2011). The two main factors that contributed to this variability are the late date of composition of the formularies and the varying degrees of theological reflection and devotional sensibility on purgatorial doctrine after the seventh century. The need to provide a proper formulary, in fact, stemmed from the redefinition of purgatorial doctrine that was mostly influenced by Bede the Venerable and other writers active between the end of the seventh and the eighth centuries (Moreira 2010, pp. 109-111; Moreira and Toscano 2010; Paxton 1990).1 Given the chronological proximity of this developments with the process of transmission of Roman chant to Francia in the eighth century, Roman chant books did not include provisions for the Dead and therefore Frankish and Roman cantors had to work independently at the creation of the mass formularies after the end of the eighth century. Beneventan manuscripts partake to this diversified tradition in that they present a mixture of chants with various origins and dates. Geographical distribution of manuscript concordances for these chants is multidirectional. Some pieces, especially those with Roman concordances, were stylistically and textually reworked, thus redefining patterns of stylistic manipulation for late chants. Each of the Beneventan manuscripts shows a distinctive blend of received, yet manipulated, super-regional chants with other less widespread compositions. The special ways in which chants are grouped allows us to identify the scribes’ organizational preferences and reveal exclusive manuscript concordances that would otherwise be difficult to detect. An example of how organizational preferences can guide us in identifying concordances that would be otherwise difficult to detect is the strict order correspondences between I-BV 39 and the gradual fragment now sewn as fol. 96 in I-BV 66 (table 1).2 These exclusive

This essay is part of a larger and collaborative project undertaken with Rebecca Maloy on the masses for the Dead in medieval manuscripts. While our project is on the interconnections between the Roman and Gregorian formularies for the Dead and the documentary value of the respective manuscript traditions, this study focuses on the chants included in Beneventan manuscripts. Part of the content of this essay is also included in my book Interlacing Traditions: Neo-Gregorian Chants Propers in Beneventan Manuscripts (Nardini 2016). This essay is dedicated to the memory of Fathers André Thibaut and Jean Mallet. Their unfailing generosity and good humor helped shaping my attitude toward research during the initial stages of my career and taught me that humbleness and inquisitiveness are necessary means for discovery.

1. The classical study on Purgatory, whose main claims are now being revisited in several studies, is Le Goff (1984).

2. I-BV 66, fol. 96 is a fragment of an eleventh-century gradual that was sewn at the end of I-BV 66, a martyrology of the twelfth century. These correspondences were discussed in my paper ‘Between St Peter and Santa Sofia: Liturgical and Notational Features in Benevento, Biblioteca Capitolare, 39’ at the International Medieval and Renaissance Conference ‘Med&Ren’ (Certaldo, 4-7 July 2013). The content of tabl 1 is also included in Nardini (2016, pp. 56-58 and 99-100) where I also discuss issues related to the manuscripts provenance.

481 CANTUS PLANUS 2014 NARDINI — THE MASSES FOR THE DEAD IN BENEVENTAN MANUSCRIPTS manuscript concordances are the possible signs that the two manuscripts shared the for communions, as the example of I-BV 39 in table 3, p. 484, shows.4 Arrangement same institutional destination, that of the monastery of Saint Peter inside the Walls. variability, a variability that is common to most European sources (Grössinger 1957),5 While further reasons for this hypothesis are discussed in my Interlacing Traditions implies a late date of composition, but also that the chants could possibly be assembled, (Nardini 2016), here I would like to underline that no other Beneventan manuscript originally, in separate libelli in which chants were grouped by liturgical genre and not displays the same order as I-BV 39 and I-BV 66, fol. 96 for these formularies. by formulary. The existence of libelli has been often suggested in chant studies and seems to be confirmed by other sanctoral feasts. The late-eleventh century inventory Table 1. Order of chants in I-BV 66, and I-BV 39 of San Pietro a Patierno near Naples, published by Domenico Mallardo (1952), for instance, lists a number of «guaderni» devoted to masses and offices for individual

I-BV 66 I-BV 39 feasts (Nardini, 2015 and 2016). In the case of the Dead, the tradition of separate books is also confirmed by the later manuscript tradition. Individual manuscripts were [Gr Qui Lazarum resuscitasti Gr Qui Lazarum resuscitasti often fully devoted to the incorporation of chants and prayers for mass and office, Re]quiem eternam Requiem eternam a circumstance also prompted by the need to facilitate the celebration of rituals in Tr De profundis * variable locations (churches, cemeteries, private homes, and so on).6 Of Domine convertere * Table 2. Missa pro defunctis, I-BV 34, fols. 265v-266v Of Subvenite sancti Of Subvenite sancti

Co Chorus angelorum Co Chorus angelorum In Requiem eternam al mis defun[ Al Ps De profundis Convertere Quia eripuit In Si enim credimus In Si enim credimus Gr / Ps De profundis Ps De profundis Gr Requiem eternam / Anime eorum V Et sicut in Adam Gr Requie[m eternam / Anime eorum] Gr Requiem eternam / Anime eorum Tr De profundis* Of Domine Iesu Xpiste Of Domine fac mecum * Of Domine fa mecum * V Hostias et preces Of Domine Hiesu Christe Of Domine Hiesu Christe V Hostias et preces V Hostias et preces Of Domine convertere *

Co Donet eis dominus … [ Co Donet eis dominus Co Donet eis dominus

Co Lux eterna Co Omne quod dat

Co Ego sum resurrection All chants are notated. Asterisks indicate truncated chants Co Lux eternam

Among Beneventan manuscripts, I-BV 34 (table 2) is the only one in which mass All chants are notated. Asterisks indicate truncated chants propers are arranged in a sort of commune defunctorum, where chants are placed after the single rubric of missa pro defunctis in the liturgical order of mass genres.3 With 4. Missals, like V-CVbav lat 7231 from Apulia or I-MC 127 from Montecassino provide details about the specific locations one introit, two graduals, one tract, two offertories, and four communions this is the for mass celebration (in the church, at the cemetery, and so on), and the various suffrage occasions after the decease. The complete formularies from the manuscripts of the Biblioteca Capitolare in Benevento are also published in Mallet and manuscript with the smallest provision for the Dead. All other manuscripts organize Thibaut (1997, 2, pp. 660-663). the chants into multiple formularies, often providing alternative pieces, especially 5. For more details see Nardini (2016, pp. 153-156).

6. Separate books or libelli for the mass and office of the Dead were very common during the Middle Ages and the early modern era, as several extant exemplars and inventories testify. One such example comes from the church of San Giacomo degl’Italiani in Naples, whose inventories mention ‘messali’ or ‘messaletti per la Messa dei defunti’, Naples, Archivio Storico 3. The manuscript is published in facsimile (Le Codex VI. 34, 1937). Diocesano, Fondo inventari, I -Nasd, VI-83; III-20.

482 483 CANTUS PLANUS 2014 NARDINI — THE MASSES FOR THE DEAD IN BENEVENTAN MANUSCRIPTS

Table 3. Mass formularies for the Dead in I-BV 39 for wave-like melodic contours and scalar melodic segments that are typical of neo- Gregorian chants in southern Italy (Nardini 2006). Qui Lazarum is fully gregorianized Missa pro defunctis in non Roman manuscripts and follows the model of the Gregorian Beata gens (Nardini In Requiem eternam 2016). Table 4 shows a complete comparison between I-BV 39, V-CVbav F 11, and Gr Convertere V-CVbav lat 5319. All Requiem eternam Table 4. Comparison between I-BV 39, V-CVbav F 11, and V-CVbav lat 5319 Of Domine convertere Co Omne quod dat I-BV 39 V-CVbav F 11 V-CVbav lat 5319 Co Ego sum resurrection In Rogamus te domine In Rogamus te domine In Rogamus te domine Gr Qui Lazarum Gr Qui Lazarum Gr Qui Lazarum Alia Tr De profundis Tr De profundis Tr De profundis In Rogamus te domine Of Domine convertere Of Domine convertere Of Domine convertere Gr Qui Lazarum Of Subvenite sancti Of Domine Hiesu Tr De profundis Co Chorus angelorum Co Lux eterna […] Co Lux eterna […] Of Domine convertere quia pius est (melody 1) quia pius es (melody 1) Of Subvenite sancti Co Lux eternal […] quia pius es (melody 2) Co Chorus angelorum Co Christus qui natus

Alia

In Si enim credimus As mentioned, the first part of the formulary is practically the same in all manuscripts, Gr Requiem eternam while the choice and order of offertories and especially of communions follow a Of Domine fac mecum more unpredictable order. The other two formularies, Requiem eternam and Si enim Of Domine Hiesu Christe credimus, could be of Frankish origins, although the communion Lux eterna could be a Roman piece composed after the transmission of the core repertory to Francia. Co Donet eis dominus Reasons for inferring a Roman origin for this communion include the melodic Co Lux eterna similarity with other Roman pieces, such as the first part of the communionSpiritus sanctus for the second feria after Pentecost, and some ornamentational features that are unusual for Gregorian chants that I have discussed elsewhere (Maloy and Nardini With the exception of I-BV 34, and similarly to I-BV 39, all other manuscripts arrange 2013; Nardini, 2016). Manuscript comparison, in fact, highlights a very unusual the repertory in three formularies: Requiem eternam, Rogamus te domine, and Si type of melodic instability for this piece in that ornamentations always occur in the enim credimus. In these formularies there is a mix of Roman and neo-Gregorian guise of pitch ornamentation and are constantly placed on the same syllables («e-is» compositions. The formulary Rogamus for the first part mostly follows the order and «e-ternum»). In addition, they defy any principles of regional continuity. There of Roman manuscripts. However the Beneventan versions of both Rogamus and is a group of neo-Gregorian communions that display a similar type of variants in Qui Lazarum are melodically different from those found in Roman manuscripts, melodic ornamentation in the versions copied in F-CHRm 47 and in some Beneventan as demonstrated by Rebecca Maloy (Maloy and Nardini 2013).7 They both have manuscripts.8 For Lux eterna I-BV 34 and I-BV 39, the two Beneventan manuscripts concordances outside of Benevento, mostly in Italian manuscripts and in F-CHRm transmitting the melody for this communion,9 while sharing the same version, still 47. The Beneventan version of Rogamus belongs to the largest group of introits based present variants from other regional versions on «eis» (example 1, p. 486). on the same tune family found in Beneventan manuscripts. It displays a preference

8. Examples of these variants are included in Nardini (2016, pp. 357-361).

7. These chants are also discussed in Nardini (2016, transcription on p. 195). 9. V-CVbav lat 7231, fol. 36v, only has the text with no melody.

484 485 CANTUS PLANUS 2014 NARDINI — THE MASSES FOR THE DEAD IN BENEVENTAN MANUSCRIPTS

Example 1. Melodic comparison between I-BV 34 and I-BV 39 Example 2. Melodic and textual comparison for Pro quorum memoria in selected manuscripts

I-Ps 120 I-BV 34, fol. 266v fol. 131v

I-MOd 7 fol. 187v I-BV 39, fol. 195r

I-Pc 47 fol. 226

The communion Pro quorum memoria, also found in the Roman manuscript V-CVbav F11 for the formulary Rogamus te domine and in I-BV 35 for the formulary Si enim I-BV 35 fol. 170 credimus (V-VCbav lat 7231 has a different arrangement, as mentioned above), presents a different manner of formal manipulation. The piece is widely spread in European sources with several textual and melodic variants. Partial manuscript comparison for V-CVbav 7231 this communion shows that, besides a common upward opening gesture, melodic fol. 37 versions vary greatly from source to source (example 2). In I-Ps 120, as noticed by Rebecca Maloy, Pro quorum memoria appears as the continuation of Christus qui natus, V-CVbav F 11 another communion with Roman concordances. It is possible that the conflation of the fol. 68v two communions was a scribal mistake prompted by a non-rubricated exemplar.10 Other versions present melodic transpositions and varying ornamentations. The version of V-CVbav F11, despite textual variants, is very similar to that of I-MOd 7 up to I-Ps 120 11 «eis domine». Also worth of notice is that I-MOd7 and I-Pc47, two sources sharing the fol. 131v same regional origin, have varying melodies.12 This provides another example of non- regional continuity of melodic variants for chants belonging to the mass of the Dead, as in the case of Lux eterna mentioned above. I-MOd 7 fol. 187v

The piece does not belong to the type of antiphon- or responsory-communions considered by McKinnon and Maiani as belonging to a more recent layer of Roman I-Pc 47 communions (McKinnon 1992 and 2000, pp. 336-339; Maiani 1996 and 1998). There fol. 226 are several antiphons, however, with a melody that is very similar to the version found in the Beneventan manuscripts and one antiphon with a melody that is similar, at least I-BV 35 for the first part, to the melody found in V-CVbav F 11 and I-MOd 7, as a search on the fol. 170 Cantus Manuscript Database allowed to determine.13

V-CVbav 7231 fol. 37 10. Other cases of non rubricated communions that could have prompted similar scribal mistakes were discussed in Maloy and Nardini (2013) and will appear in our forthcoming essay.

11. Textual variants are discussed in more details below.

12. The chant in I-Pc47 is transposed a fifth below and presents textual and melodic variants from I-MOd 7, especially on V-CVbav F 11 «quo-rum», «haec hostias offerimus/a corpus Christi sumimus», and on «requiem sem-piterna/m». fol. 68v

13. http://cantus.uwaterloo.ca/search?t=beata+caecilia.

486 487 CANTUS PLANUS 2014 NARDINI — THE MASSES FOR THE DEAD IN BENEVENTAN MANUSCRIPTS

Example 3 shows a comparison between the communion in V-CVbav F 11/I-MOd 7 Example 4 shows an even more striking similarity between the version of I-BV 35 and the antiphon Beata Cecilia as found in the fourteenth-century antiphoner from the and V-CVbav lat 7231 of Pro quorum and the antiphon Cosmas et Damianus, as found Abbey of Sankt Lambrecht (Steiermark, Austria), A-Gu 30, available through the Cantus in I-BV 21.15 There are various antiphons with this melody, including the Solesmes Database.14 The first phrase has the same profile in all versions, notwithstanding the contrafactum Descendit Jesus cum Maria (LU 1439).16 Cosmas et Damianus, however, textual differences between I-MOd 7 and V-CVbav F 11 mentioned above. The second appears as the one that is the closest melodically to the Beneventan version of Pro phrase, while presenting a recitation on C in A-Gu 30 on the words «hodie te fateor […] quorum memoriam. Only variants are the upward leap on «Anthemus» in the antiphon, quia» that is absent in the communion, proceeds in both the antiphon and communion while the communion lingers on a and g on «noto-riam», the stepwise motion on «co- with a similar melodic contour with the most ornamented version on «requi-em sem- ronati» in the antiphon, as opposed to the back and forth leap of the communion on pi-ternam» in I-MOd 7. «tu-is», and the use of the Gallican cadence in the communion.17

Example 3. Melodic comparison between communion Pro quorum memoria and Beata Cecilia Example 4. Comparison between the communion Pro quorum memoria 1. I-MOd 7, fol. 187v; and the antiphon Cosmas et Damianus 2. V-CVbav F 11, fol. 68v; 1. I-BV 35, fol. 170r-v 3. A-GU 30, fol. 310 2. I-BV 21, fol. 247v

Certainly more research is needed to determine possible directions of influence for this communion. There are two elements that are important to underline about the melodic correspondences identified thus far. First, the melodic similarity between some versions of this communion and office antiphons points to the circumstance

15. Le Codex 21, 2001.

16. I thank Luca Ricossa for providing some additional melodic concordances for the antiphons Illuminare Domine, Adaperiat Dominus, and Veni sponsa Christi, all of whom, while sharing the same melodic profile, are not as close to the Beneventan version of Pro quorum as Cosmas et Damianus.

14. http://www.literature.at/viewer.alo?objid=1141&viewmode=fullscreen&scale=2&rotate=&page=620. 17. The version of Cosmas et Damianus found in A-GU 30, fol. 260 also closes with a Gallican cadence.

488 489 CANTUS PLANUS 2014 NARDINI — THE MASSES FOR THE DEAD IN BENEVENTAN MANUSCRIPTS that communions and antiphons could share melodies even when their texts were most immediate extant copy is that of Sant’Angelo in Formis near Capua (Lavin 1990; different. Second, the melodic similarity between I-MOd 7 and V-CVbav F 11 points Tronzo 1985).20 While those depictions underlined the sense of awe for the glory of to possible exclusive exchanges between Rome and Ravenna. In other works I have God and the despair for his eternal punishment and were meant as deterrents for highlighted that repertorial and stylistic similarities between Rome and Benevento a sinful life, liturgical prayers and chants for the Dead more directly focused on the can be ascribed to the strong political and pastoral influence of the Roman Curia on theme of the mercy of God. They were modeled on the Orationes Pseudocyprianeae or the Beneventan region after the end of the tenth century (Nardini 2010a and b). This on the «[p]rayers appended to Cesarius of Arles’s early sixth-century Rule for Virgin concordance between Rome and Ravenna points to other possible routes of mutual … [which] encouraged faithful Christians to celebrate death with positive images of interchanges between the Roman curia and other centers of the Italian peninsula.18 what awaited them in the next world» (Effros 2002, p. 169). This was also perceived as Example 2 also shows that manuscript versions vary not only in the melody, but also a better way to reinforce the liaison between the Dead and the Living at the moment of in the texts. I-MOd 7 is the only one with the first sentence ending with «haec hostias the farewell for and remembrance of a beloved one, a liaison that is at the core of the offerimus», while the other versions have «corpus Christi sumimus/sumitur» (I-Pc47; tradition of funerary rites and the practice of indulgence masses. V-CV-bav F 11), «frangitur» (I-BV 35), «subditur» (V-CVbav 7231). While the numerous manuscript concordances as well as the varying versions hinder the possibility of Above I have suggested that the flourishing of mass formularies for the Dead owes to finding the ultimate origins for this communion, there are some interesting aspects the development of purgatorial doctrine after the end of the eighth century. Bonnie about chant circulation and manipulation that this communion allows to highlight. Effros also discussed (2002, pp. 169-204) the possibility that the emergence of funerary First, manuscript comparison shows that for Pro quorum memoria we have, differently ceremonials, and consequently of the liturgy for the Dead, was also a means for the from what happens for Lux eterna, the other widely spread communion for the Dead, clergy to re-appropriate rites in which they did not play a determinant role until a variation of the melodic line and not the varying ornamentation of single pitches. the eighth century. For chant scholars, these masses show once again the multi- However, the variants between I-MOd 7 and I-Pc 47 show that regional continuity directionality of late chant transmission, a multi-directionality that encompasses the should not always be expected. Second, melodic concordances between the repertory city of Rome, and illuminates on patterns of melodic and textual manipulation, thus of communions and antiphons is not limited to chants sharing the same text (as in reinforcing the relevance of late chant repertories. the cases studied by McKinnon and Maiani), but involves also chants with different texts.19 Finally, mutual interchanges between the Roman curia and other centers were still happening, and even without Frankish intermediation, after the ninth century. It is not possible to ascertain, at least for now, the directions of influence, but the case of the melodic correspondences between F11 and Mod7 shows preferred routes of exchanges at least for this specific melody. Notwithstanding this highly varied tradition that attains to the selection and order of pieces as well as to the melodic and textual versions of chants, a common element in the mass formularies for the Dead is that all texts are characterized by generically positive references to the eternal rest and quiet of the soul. Other scholars have underlined that the development of purgatorial doctrine in the eighth century, while playing a role in the creation of ritual formularies, did not influence the content of liturgical texts. Medieval believers were well aware of the pains of divine punishment through the reading of theological writings or, for the illiterates, the viewing of church depictions. Well known is the tradition of decorating scenes of the Apocalypse and the Last Judgment on the counter-façade of important churches, as for instance those of the Santi Maria and Donato in Torcello or the much later Cappella degli Scrovegni in Padova, all stemming from the tradition of the now lost pictorial cycle of the Constantinian basilica of the Old Saint Peter’s in Rome, whose

18. The connections between Rome and other centers are also discussed in Dyer (2014) and in Nardini (2010b).

19. This assertion would require a broader and more systematic melodic comparison between communions and antiphons 20. An overview of the theme of the Last Judgment in churches from Late Antiquity to the later Middle Ages can be found or responsories, a comparison that is now more easily feasible thanks to the Cantus and Global Chant databases. in Laffin (2009).

490 491 CANTUS PLANUS 2014 NARDINI — THE MASSES FOR THE DEAD IN BENEVENTAN MANUSCRIPTS

Manuscript cited Bibliography

A-GU 30 Graz, Universitätsbibliothek, 30 (olim 38/9 f.) Dyer Joseph (2014), St. Peter and His Neighbors: Reflections on Roman and Italian Chant and Liturgy Antiphoner, St-Lambrecht, 1300s in Frank Hentschel - Winkelmüller eds., Nationes, Gentes und die Musik im Mittelalter, 287-340, Berlin-Boston: Gruyter. F-CHRm 47 Chartres, Bibliothèque Capitulaire, 47 Gradual, Bretagne, ninth-tenth century Effros Bonnie (2002), Caring for Body and Soul: Burial and the Afterlife in the Merovingian World, University Park, PA: The Pennsylvania State University Press. I-BV 21 Benevento, Biblioteca Capitolare, 21 Antiphoner, Benevento, twelfth-thirteenth century Gay Claude (1957), ‘Formulaires anciens pour la messe de défunts’, Études Grégoriennes 2, pp. 83-129. I-BV 34 Benevento, Biblioteca Capitolare, 34 Gradual with tropes and prosulas, Benevento, first half of twelfth century Grössinger Christa (2007), The Road to Hell in Peregrine Horden ed., Proceedings of the 2003 Harlaxton Symposium, Harlaxton: Tyas, pp. 318-330. I-BV 39 Benevento, Biblioteca Capitolare, 39 Gradual with tropes and proses, Benevento, Saint Peter inside the walls Hesbert René-Jean (1952), Les pièces de chant des messes «Pro Defunctis» dans la tradition manuscrite (?), end of eleventh century in Higini Anglés ed., Atti del congresso internazionale di musica sacra organizzato dal Pontificio Istituto di Musica Sacra e dalla Commissione di musica sacra per l’Anno Santo, 1950, Tournai: Desclée, I-BV 66, fol. 96 Benevento, Biblioteca Capitolare, 66, fol. 96 pp. 223-228. Fragment of a Gradual, Benevento, eleventh century Laffin Josephine (2009), ‘What Happened to the Last Judgement in the Early Church?’, Studies in Church I-MC 127 Montecassino, Archivio dell’Abbazia, 127 History 45, pp. 20-30. Missal, Montecassino, eleventh-twelfth century I-MOd 7 Modena, Biblioteca Capitolare, O.I.7 Lavin Marylin Aronberg (1990), The Place of Narrative: Mural Decoration in Italian Churches, 431-1600, Gradual with tropes and proses, Forlimpopoli, late eleventh century Chicago: University of Chicago Press. I-Pc 47 Padua, Biblioteca Capitolare, 47 Le Codex 21 de la Bibliothèque Capitulaire de Bénévent: Antiphonale monasticum (XI-XIIIe S.) (2001), Gradual with tropes and proses. Ravenna, twelfth century (PalMus 22).

I-Ps 120 Pistoia, Biblioteca Capitolare, 120 Le Codex VI. 34 de la Bibliothèque Capitulaire de Bénévent (XIe-XIIe siècle): Graduel de Bénévent avec prosaire Gradual with tropes. Pistoia (?), beginning of twelfth century et tropaire (1937), (PalMus 15). V-CVbav lat 7231 Vatican City, Biblioteca Apostolica Vaticana, Vat. Lat. 7231 Le Goff Jacques (1984), The Birth of Purgatory, Eng. trans. by Arthur Goldhammer, Chicago: University of Apulia/Dalmatia, thirteenth century Chicago Press; or. ed. La Naissance du Purgatoire, 1981, Paris: Gallimard. V-CVbav F 11 Vatican City, Biblioteca Apostolica Vaticana, Vat. lat. F 11 Ritual-orationale, Rome, twelfth century Maiani, Bradford C. (1996), The Responsory-Communions: Toward a Chronology of Selected Proper Chants, Ph. D. dissertation, University of North Carolina at Chapell Hill. V-CVbav lat 5319 Vatican City, Biblioteca Apostolica Vaticana, Vat. lat. 5319 Gradual with Roman chant-Kyriale-Proser, Rome, – (1998) ‘The Responsory-Communion for Paschaltide’, Studia Musicologica Academiae Scientiarum San Giovanni in Laterano (?), late eleventh–beginning of twelfth century Hungaricae 39, pp. 233-240.

Mallet Jean - Thibaut André (1997), Les manuscrits en écriture bénéventaine de la Bibliothèque capitulaire de Bénévent, 2 vols., Ottawa: Institute of Mediæval Music - Paris: Cnrs - Turnhaout: Brepols.

Maloy, Rebecca - Nardini Luisa (2013), ‘Musical Hybridization in the Roman Mass for the Dead’, Paper read at the meeting of the American Musicological Society, Pittsburgh, 2013.

McKinnon, James (1992), ‘The Roman Post-Pentecostal Communion Series’ in László Dobszay ed., Cantus Planus: Papers Read at the Fourth Meeting, Pécs, Hungary, September 1990, Budapest, Hungarian Academy of Sciences, pp. 175-186.

– (2000), The Advent Project: The Later-Seventh-Century Creation of the Roman Mass Proper, Berkeley, CA: University of California Press.

Moreira Isabel (2010), Heaven’s Purge: Purgatory in Late Antiquity, Oxford-New York: Oxford University Press.

Moreira, Isabel - Toscano Margaret eds. (2010), Hell and Its Afterlife: Historical and Contemporary Perspectives, Farnham: Ashgate.

492 493 CANTUS PLANUS 2014

Matthew Peattie – (2015), The Masses for the Holy Cross in Some Italian Manuscripts in Stanley Boorman and Anna Zayaruznaya eds., «Qui musicam in se habet»: Essays in Honor of Alejandro Planchart, Madison (Wi): American Institute of Musicology, pp. 41-70. Graphic difference and the interpretation

Nardini Luisa (2006), Non-Standard Introits in Beneventan Manuscripts: Compositional Strategies and of the climacus in Beneventan notation Transmission in the Sources in László Dobszay ed., Cantus Planus: Papers Read at the 12th Meeting of the IMS Study Group, Lillafüred, Hungary, 2004, Budapest: Institute for Musicology of the Hungarian Academy of Sciences, pp. 345-369.

– (2010a) ‘Roman Intruders in non-Roman Chant Manuscripts: The Cases of «Sint lumbi vestri» and «Domine si tu es»’, Acta Musicologica 82, pp. 1–20. Eleventh-century sources in Beneventan notation preserve at least eight distinct ways of notating the climacus. In this paper, I demonstrate that there are important – (2010b) ‘The St Peter Connection and the Acquisition of a Roman Offertory in Bologna and Benevento’, Mediaeval Studies 82, pp. 39–74. functional differences between these graphic forms, and argue that the differences have implications for transcription and performance. Although this study focuses on graphic – (2016), Interlacing Traditions: Neo-Gregorian Chant Propers in Beneventan Manuscripts, Toronto: Pontifical Institute of Mediaeval Studies. difference in the notation of a single figure, it introduces comparative approaches that set the stage for a comprehensive consideration of nuances of intensity and duration – (2017) In the Quest of Gallican Remnants in Gregorian Manuscripts: Archaisms in the Masses for the in south Italian manuscripts. This paper focuses primarily on I-BV 40, a mid-eleventh- Holy Cross in Aquitanian Chant Books in Patricia Hall ed., The Oxford Handbook of Music Censorship, Oxford: Oxford University Press, pp. 7-38. century gradual from Benevento (henceforth BV 40). This manuscript is an important repository of the south Italian recension of the Gregorian repertory, as well as the axton P Frederick S. (1990), Christianizing Death: The Creation of a Ritual Process in Early Medieval Europe, 1 Ithaca: Cornell University Press. central surviving source of the music of the Beneventan rite. The two repertories allow for contrasting approaches to comparative study. Examples from the Beneventan rite Sicard, Damien (1978), La liturgie de la mort dans l’Église Latine des Origines à la Reforme Carolingienne, are suited to the study of melodic context and the comparison of variant neumations in Münster: Aschendorffsche Verlagsbuchhandlung. a small number of concordant sources. Examples from the Gregorian repertory allow a Tronzo William (1985), The Prestige of Saint Peter’s: Observations on the Function of Monumental Narrative broader comparative framework: in this paper, I consider the neumations of Gregorian Cycles in Italy in Herbert L. Kessler - Marianna Shreve Simpson eds., Pictorial Narrative in Antiquity and the Middle Ages, Washington, DC: National Gallery of Art. music in BV 40 alongside nuance rich sources of the central repertory, an approach that brings up important questions about the relationship between the central and south Valle Nemesio (2011), A Coalescence of Liturgical Consensus on the Chants for the Mass for the Dead from Italian recensions of the Gregorian repertory, to which I will return below. Its Origins through the Fourteenth Century, Ph. D. dissertation, Pittsburgh: University of Pittsburgh.

Eight of the most common forms of the climacus in BV 40 are illustrated in table 1, p. 496.2 My first observations concern melodic context. The neumes numbered 1 and 2 show two graphically distinct ways of writing a climacus that begins at the same pitch as the preceding note. The next two figures show different ways of writing a climacus that begins higher than the previous note. No. 5 is used exclusively when the initial pitch of the climacus is lower than the note before. Nos. 6-8, which employ special symbols, appear in a variety of contexts.

1. A colour facsimile of BV 40 is printed as Benevento, Biblioteca Capitolare 40. Graduale (Albarosa and Turco 1991). The indices separate the music of the Beneventan rite (referred to here as the Beneventan repertory) from the rest of the contents, gathered under the rubric «repertorio gregoriano». The latter corpus includes the south Italian recension of the central repertory, as well as many melodies in Gregorian style with local or pan-Italian transmission. For the purposes of this study, I limit the consideration of ‘Gregorian’ music in BV 40 to those pieces also transmitted in the seven sources represented in the Graduale Triplex (1979).

2. The Beneventan notation in these examples is printed in a typeface based on BV 40. The typeface is an editorial interpretation of the notation in graphic form in which a single glyph is used to represent a class of like symbols from the manuscript. For each class of symbols, the glyph replaces the variation of the copyist’s hand with a single schematic representation. The term ‘glyph’ is appropriated from the study of text typography, which I have explored in detail (2014). In present volume (pp. 137-142), Andreas Haug, Towards a Semiotically Informed Transcription Practice, theorizes the relationship between scribal inscriptions and characters using language derived from semiotic theory.

494 495 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Table 1. The climacus in Benevento, Biblioteca capitolare, BV 40 Appendix 1, p. 506, illustrates these correspondences in context, showing passages from BV 40 in synoptic transcription with the analogous passages from E 121, SGs 359, and graphic directionality graphic directionality LA 233. The appendices illustrate typical examples selected from a comparison of every 1. equal to the preceding note 5. lower than the preceding note melody shared by BV 40 and at least one of the northern sources.4 In appendix 1 (nos. 1-6), the figure with a horizontal uppermost stroke in BV 40, almost always corresponds to a climacus with some degree of rhythmic weight.5 The scribes of SGs 359 and E 2. equal to the preceding note 6. lower than the preceding note 121 add episemas to the uppermost notes, or notate the lower notes as tractulus; the scribe of LA 239 frequently employs a virga or uncinus as the uppermost note, and in 3. higher than the preceding note 7. equal or higher than the several instances notates uncinus for one or more of the lower notes. In nos. 2 and 6, preceding note; also in ligation the scribes of both northern traditions use heavily weighted three and four note figures. to a preceding pitch below Appendix 1 (nos. 7-12) shows something very different. The climacus with the curved 4. higher than the preceding note 8. higher or lower than the oblique uppermost stroke almost always corresponds to a neumation that is relatively preceding note light or neutral in the northern sources. All of the examples from LA 239 notate the upper two notes as dot puncta, followed by either an uncinus or a dot. The scribe of SGs 359 generally employs the simple symbol for the climacus, sometimes reinforced by the Aspects of directionality are fundamental to the calligraphy of Beneventan neumes, and adjunct letter ‘c’ (celeriter = quickly), and occasionally with a tractulus as the lowest much of the graphic variety results from different ways of writing neumes that begin note. The neumes of E 121, SGs 359, and LA 239 are described here using terminology higher, lower, or at the same height as the preceding stroke of the scribe’s pen. But derived from Cardine (1982). See especially the tables on pp. 12-15, which outline the because more than one figure is used for each melodic context, it is obvious that not simple symbols and the various modifications that represent nuance. It is important to all graphic differences can be explained by directionality alone.3 A comparison of the emphasize that the correspondence between these sources is not precise. In Benevento, two graphic forms of the climacus in which the uppermost note is equal in pitch to the distinct forms of the climacus are differentiated by the graphic shape of the uppermost the preceding note reveals a clear pattern, summarized in example 1. The figure with note. The lower notes of the three note climacus are normally notated with an oblique the horizontal uppermost stroke corresponds to passages in which the scribes of three stroke followed by a downward vertical stroke, while four note figures are usually notated important central sources of the Gregorian repertory – CH-E 121 (henceforth E 121), with a characteristic succession of oblique, horizontal and vertical strokes (example 2). CH-SGs 359 (henceforth SGs 359), F-LA (henceforth LA) 239 – indicate some degree of Example 2. The climacus in Beneventan sources emphasis or weight; the figure with the curved oblique stroke corresponds to passages in which the scribe of the northern sources employs a neutral or light neumation.

Example 1. A comparative summary of two graphic forms of the climacus in which the uppermost note is equal to the pitch of the preceding note

The consistency with which the lower notes are written suggests a scribal convention devoid of rhythmic meaning. On the other hand, the scribes of Saint Gall and Laon are able to emphasize the rhythmic weight of any of the notes. Because the Beneventan forms are differentiated only by the uppermost note, I suspect that these figures are best understood as complete musical gestures, and should not be analyzed with the same sort of note-to-note detail appropriate to the northern sources. The contrast between two graphic forms of the climacus that begin higher than the preceding pitch 3. Previous studies have noted the graphic variety of the climacus in Beneventan notation. The authors of Codex VI 34 (1937, pp. 132-133) illustrate examples from the early eleventh to twelfth century, with a focus on directionality. Rupert Fischer’s brief introduction to the notation of BV 40 (1991, p. xi) illustrates eight distinct graphic varieties of the climacus 4. The appendices list additional passages with analogous neumations below each column, and note the few exceptions in and emphasizes the need for further inquiry into aspects of rhythm. There are several graphic combinations of the climacus which there is no apparent relation between the rhythmic weight of the northern and south Italian sources. not included in this study; the scribes of the Beneventan repertory in particular notate the climacus with a clivis as the uppermost element (see for example BV 40, fol. 61, line 7 and fol. 71, lines 7 and 8). The authors of Codex VI 34 signal this 5. Here, and in the comparative examples below, I drawn on concepts of relative rhythmic weight demonstrated by Cardine figure as a rarity, and note that this form is more common in the notation of the Ambrosian rite of Milan. (1982), as well as a substantial body of subsequent literature including studies by Augustoni and Göschl (1987-1992).

496 497 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION is summarized in example 3. Again, the comparison of BV 40 with E 121, SGs 359 and In this instance, it appears that the scribe of BV 40 diverges from the ‘normal’ way of LA 239 reveals a contrast of relative weight. The climacus with the curved oblique writing the climacus to signify emphasis. This example highlights the uncertain relation stroke generally corresponds to light or neutral neumations; the climacus with the between notation and performance. In all music notation there is much that is written virga as the uppermost note generally corresponds to neumations that are, relatively and much that depends on aural understanding and conventions of performance. It is speaking, rhythmically weighted. Examples in synoptic transcription are shown in difficult to know to what degree the notation of BV 40, is descriptive or prescriptive, appendix 2, p. 508. which elements are captured precisely by the notation, and which elements simply serve to remind the singer of a gesture already ingrained in common practice. Given Example 3. A comparative summary of two graphic forms of the climacus in which that this passage represents a significant departure from graphic convention, a single the uppermost note is higher than the pitch of the preceding note strategically placed horizontal stroke may be enough to signal that the entire passage is weighted. The surrounding context reinforces this interpretation: the second climacus is written with a virga, presumably representing weight, and the neume that follows – a gradata – may also represent emphasis.6 The climacus with an s-shaped figure as the uppermost note appears relatively infrequently in BV 40. In both the Gregorian and Beneventan repertories, it usually follows a higher note, but is notated very occasionally at the unison. When the s-shaped climacus is aligned with the text, it normally falls on an unaccented syllable (I was unable to identify a single example where it was employed on an unequivocally strong syllable). It is also used in melismatic contexts, The figure on the right (appendix 2, nos. 7-12) contains several neumations that merit such as the jubilus of alleluia Iustus ut palma (ibidem, fol. 87v). Appendix 3a illustrates further comment. In no. 1, In voluntate tua, two consecutive iterations of the climacus examples of the s-shaped climacus in the Gregorian repertory. The analogous passages are supplied with contrasting graphics. The rhythmic gesture is similar to the in E 121, SGs 359 and LA 239 are always noted with lightly weighted uppermost notes; neumation of LA 239, but the relative weight is reversed in E 121. A similar contrast the typical corresponding neumations are summarized in example 4. of graphics occurs in no. 4 Constitues eius, but here, the scribes of LA 239 and SGs 359 use heavily weighted neumations for both figures. In no. 3, the scribe of Benevento Example 4. A climacus with an s-shaped uppermost note employs a virga as the uppermost note of figures, while the scribes of LA and SGs double the length of the first climacus. The juxtaposition of two distinct graphics in Benevento suggests a deliberate differentiation between these figures, but the graphic distinctions (and relative weight) correspond in varying degrees of precision to the northern sources. No. 12, from the gradual Sacerdotes eius, provides a useful window on the notation of nuance. In BV 40, the middle note of the first climacus is written with a horizontal stroke, a clear departure from the convention of notating the middle note with an oblique neume. The analogous passage in SGs 359 is notated with marked degree of rhythmic weight. These passages are shown in facsimile in figure 1. In the gradual Timete (appendix 3, no. 5, p. 510) the scribe of BV 40 writes an s-shaped climacus, while the scribe of SGs 359 uses a clivis modified by the letter ‘c’. The substitution of a clivis (presumably representing a third) is not atypical; I identified several passages in which the scribe of BV 40 writes three notes while the northern scribes write two, and while the number of notes differ, the gesture or rhythmic weight is similar. Rupert Fischer (1991, p. xi) classifies this graphic shape as an oriscus; John Boe (1992, p. 519) counters that it is unrelated to the oriscus. Neither author clearly differentiates the use of this figure in the context of the climacus from a related graphic

Figure 1 6. PalMus 15, 1937, p. 132; the authors write, pp. 124-125, that the gradata is derived from the pes inflatilis, which is Graduale Sacerdotes eius in BV 40, fol. 104v and SGs 359 (Cantatorium), p. 126 demonstrated to represent an allongement or lengthening.

498 499 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION that appears in isolation, attached to a single syllable of text. For the purposes of this Example 6. Climacus with two oblique strokes as the uppermost note in the Beneventan repertory paper, I avoid the baggage of nomenclature, and focus instead on the way this figure is used in context. Based on this study, it is not clear that this figure is analogous to any 1. Ingressa Dum sacra (BV 40, fol. 61) 2. Offertorium O quam (BV 40, fol. 133v) of the ‘special neumes’ of the northern tradition.7 In the music of the Beneventan rite, this figure is used in contexts similar to the Gregorian, where it is normally employed on a weak syllable. Selected examples from the Beneventan repertory are illustrated in appendix 3b. The scribe of BV 40 frequently notates the uppermost note of the climacus with a distinctive figure composed of two parallel oblique strokes ( ). It is used in a variety of melodic contexts. In the Gregorian repertory, it always occurs at the same pitch as the preceding note. In the Beneventan repertory, it appears with equal frequency at the ar - chan - ge - lus Mi - cha - el cum xpi - sto re - gnant same pitch as, or higher than, the preceding note. In the Gregorian repertory, this form of the climacus almost always corresponds to weighted neumations in E 121, SGs 359 3. Communio Ad honorem (BV 40, fol. 134) 4. Communio Quid ad nos (BV 40, fol. 143) and LA 239. When this figure is employed as the uppermost note of a four note climacus in Benevento, the two upper notes receive emphasis in the northern sources. The typical corresponding neumations are illustrated in example 5; synoptic transcriptions are illustrated in appendix 4, p. 512. The uppermost note of this figure is sometimes ligated to the descending shaft of a neume that precedes it. In alleluia Exultate deo (appendix 4, no. 12) it corresponds to a weighted neumation in LA 239 and SGs 359. In alleluia Te et ec - ce cre - de - re de - si - de - ras decet ymnus (appendix 4, no. 13), the uppermost note of the climacus is unweighted, but the preceding note (at the unison) is modified in LA 239 by addition of the adjunct letter 5. Communio Gaudent (BV 40, fol. 133) 6. Communio Zacharias pater (BV 40, fol. 89v) ‘t’ (tenere = to hold), and in SGs 359 by the addition of an episema.

Example 5. A climacus with two obliques strokes as the uppermost note in the Gregorian repertory

- a pro e - ius a - mo - re vi - si - ta - vit et fe -

7. Responsorium Tenebre (BV 40, fol. 14)

A few additional examples from the Beneventan repertory are illustrated in example 6. In the Beneventan repertory, this figure is frequently, but not always, employed de - re - li - qui - sti? Et in - cli - na - to on strong syllables, and is used pervasively in several of the most common cadence formulas. In the Beneventan repertory it is frequently ligated to a lower note that 8. Ingressa Michi autem nimis (BV 40, fol. 71v) precedes it, a graphic from that not used in the notation of the central Gregorian repertory in BV 40.

7. The authors of PalMus 15, 1937 describe this figure in the context of the climacus as «trait oblique accompagné d’un poser et lâcher de plume». They note that the s-shaped figure frequently corresponds to Do or Fa; in BV 40, this observation a - mi - ci tu - i de - i appears frequently to be true in the Gregorian repertory, but not in Beneventan.

500 501 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Although we can discern patterns of relative weight between the climacus with an Communio Qui mandu- (BV 40, fol. 28) Offertorium Milia milium (BV 40, fol. 61) s-shaped uppermost neume, and the climacus with a parallel oblique uppermost neume, the comparisons do not adequately explain why the scribe uses special graphic shapes when relative importance alone could be illustrated simply by using the horizontal, oblique or vertical figures illustrated above. The key to the differences may lie in the san-gui - et de - ways that these two forms are contrasted. These figures are frequently used in close proximity – perhaps representing alternating patterns of gesture that are relatively Communio Qui mandu- (BV 38, fol. 53) Offertorium Milia milium (BV 38, fol. 83) strong and weak, perhaps in opposition to each other. Example 7 shows the two figures used consecutively in a passage from the Easter ingressa of the Beneventan rite.

Example 7. Two special forms of the climacus justaposed in context (BV 40, fol. 159v)

san-gui - et de -

Canticum Benedictus es (BV 40, fol. 12v) Offertorium Petrus apostolos (BV 40) - le - lu - ia

Much remains to be discovered about these figures. Both graphics are also used in isolation, in the context of compound neumes, and in various ligated forms; these [nostro]-rum do - ctor gen - ti - um figures require a comprehensive study, of which their position as the uppermost note of the climacus is only one part. A final graphic form of the climacus is illustrated in Canticum Benedictus (BV 33, fol. 71) Offertorium Petrus apostolos example 8. In this figure, the uppermost stroke is elongated, drawn as an undulating (V-CVbav Ottob. lat. 145, fol. 124v) horizontal stroke. In melodies from the Beneventan repertory shared by BV 40 and the roughly contemporaneous I-BV 38, I identified eighty-two examples of this form of the climacus; seventy-six of them receive identical neumations, indicating a stability of transmission and neumation in the notation of the Beneventan rite. The few exceptions [nostro]-rum do - ctor gen - ti - um are illustrated in example 8; in each of these exceptional examples, the analogous neume in the concordant source is notated as two distinct notes at the unison.

Example 8. The climacus with an elongate upper stroke. The alternate neumations indicate that the elongate stroke is formed by the ligation of two notes at the unison Ingressa Ecce sedet (BV 40, fol. 71) Ingressa Lumen quod (BV 40, fol. 89)

Ingressa Ecce sedet (BV 40, fol. 71) Antiphona Ad vesperum (BV 40, fol. 15v)

in me - di - o e - ius mi - ra - bi - lis

Ec-ce se - Ad ves - pe - rum Ingressa Ecce sedet (BV 38, fol. 93) Ingressa Lumen quod (BV 38, fol. 110) Ingressa Ecce sedet (BV 38, fol. 93) Antiphona Ad vesperum (BV 38, fol. 43v)

in me - di - o e - ius mi - ra - bi - lis Ec-ce se - Ad ves - pe - rum

502 503 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

This form of the climacus is rarely used in the Gregorian repertory. I identified only five With the exception of the elongated figure, which represents the ligation of two examples in pieces that are also transmitted in E 121, SGs 359, and LA 239; several of notes, it is difficult to know whether the relative emphasis represented by the distinct these passages confirm that the elongated figure represents two consecutive notes at the graphic varieties of the climacus is one of intensity or duration, whether it effects only same pitch (see, for example, alleluia: Omnes gentes plaudite at «iubilate deo» (BV 40, the uppermost note of the figure, or signals an unwritten performance practice. The fol. 66v; F-LA 174; CH-E 345). This neume appears to notate a phenomenon of ligation rhythmic nuances of the climacus outlined here are summarized in appendix 5, p. 514. at the unison particularly characteristic of south Italian music. It is used pervasively in It is clear that south Italian scribes have different ways of writing nuances of duration the Beneventan repertory, and features prominently in several characteristic cadence and intensity. John Boe (1983, pp. 43-66) has shown that the ‘acuasta’ represents a light figures, and in a common mid-phrase figure that turns around two consecutive climacus. or quick approach to unaccented syllables, equivalent to the dot punctum of Saint Gall. It occurs most frequently on recurrent reference tones characteristic of Beneventan René-Jean Herberts’s work on the pes ‘inflatilia’ demonstrates the frequency with which modality. Some typical uses of this figure are illustrated in example 9. this special form of the pes corresponds to the pes quadratus of Saint Gall (PalMus 15, pp. 124-125). The different graphic varieties of the climacus provide further evidence Example 9. Typical uses of the climacus with an elongate uppermost note in the music of the Beneventan rite that Beneventan scribes provided a record of the relative weight or importance of some notes. A summary of this study is illustrated in appendix 5. The recognition of Ingressa Ecce sedet (BV 40, fol. 71) rhythmic nuance in south Italian notation of the eleventh century has implications for performance and editing, and introduces significant questions about the relation between the central and south Italian recensions of the Gregorian repertory. It is clear that the rhythmic nuances of the northern sources are more detailed, and it is not obvious that study of Beneventan sources will reveal much about the practical interpretation of the rich nuances of Saint Gall and Laon. But in the case of repertories unique to Italian so - nat an - te e - um sources, Beneventan neumes provide an indispensible record of relative importance and rhythmic weight. We know that the scribes had the tools, and presumably thought Ingressa Postquam (BV 40, fol. 4v) it important, to notate these aspects; the graphic details that represent relative weight should be considered integral to the notation, and any method of transcription – whether in modern noteheads, or a typeface of neumatic notation – needs a consistent way to capture these details. A larger question concerns the transmission of details of performance. Although this study draws on examples from a manuscript copied in the mid-eleventh century, the graphic distinctions, and the details of nuance they represent, di - sci - pu - lo - rum su - o - rum di - are present in the earliest extant examples of music writing from the Beneventan zone.8 Since the first half of the twentieth century, scholars have remarked on aspects of Offertorium Salve crux (BV 40, fol. 142v) antiquity in the south Italian recension of the Gregorian repertory, and demonstrated liturgical and musical features that point to an early transmission of the repertory to the south. Looking at this question through the specific lens of nuance, it is apparent that there are significant moments of concordance between the central and south Italian traditions. Although the details are not as clearly delineated in Benevento, the moments of emphasis are the same; the graphic distinctions in the notation of the climacus in - pe me et red - de me ma - gi - stro BV 40 correspond to passages in the northern sources that by all appearances outline equivalent rhythmic gestures. These concordances suggests that in whatever form Communio Quis te (BV 38, fol. 38) the Gregorian repertory arrived in southern Italy, it was transmitted with a common central understanding of nuance, which came to be represented in its own way in the characteristic and stylized habits of the south Italian scribes.

8. A comparison of selected pieces in I-BV 33 and V-CVbav Vat. Lat. 10673 (both late tenth or early eleventh century) with BV 40 reveals a stability of neumation, and a use of each of the varieties of the climacus documented here. (See one graphic - num me - di - ta - re - ris example from I-BV 33 in example 8, Benedictus es, above).

504 505 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Appendix 1. Two graphic forms of the climacus in which the uppermost note is equal to the pitch of the preceding note

1. Co Cantabo domino 2. Co Ego clamavi quoniam 7. Gr Domine dominus noster V. Quoniam 8. Gr In omnem terram

LA 153 LA 133 LA 149 LA 150

SGs 141 SGs 122 E 315 E 316

BV 40 BV 40 BV 40 BV 40 fol. 143v fol. 152v fol. 145 fol. 91v

Can-ta-bo do - mi - no - di ver - ba me - a [est] ma - gni - ter - ram

3. Int Cibavit eos 4. Gr Ego dixi domine 9. Gr In omnem terram 10. Gr Ne avertas faciam tuam

LA 126 LA 149 LA 133 LA 95

E 257 SGs 139 SGs 122 SGs 94

BV 40 fol. 80 BV 40 BV 40 BV 40 fol. 143 fol. 91v fol. 2v

al - le - [ma] - la - [co] - rum Tu - am

5. Co Dico autem vobis 6. Gr Nimis honorati sunt V. Dinumerabo 11. Gr Nimis honorati sunt 12. Gr Propter veritatem

LA 146 LA 139 LA 146

E 290 SGs 136 SGs 130 SGs 136

BV 40 BV 40 BV 40 BV 40 fol. 113 fol. 127v fol. 127 fol. 115

Di - co au - tem vo - bis di - nu-me-ra-bo e - Ni - mis ho - no - ra - [iu] - sti -

506 507 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Appendix 2. Two graphic forms of the climacus in which the uppermost note is higher than the pitch of the preceding note

1. In Deus in loco sancto 2. In Dispersit 7. In In voluntate tua 8. Co Data es mihi omnis

LA 161 LA 110 LA 153 LA (lacuna)

E 322 E 286 E 336 E 222

BV 40 BV 40 BV 40 BV 40 fol. 150 fol. 37 fol. 145v fol. 109v fa - cis di - sper - sit tu es Bap - ti - zan - tes e - os

3. In Cognovi domine 4. In Dominus illuminatio 9. Gr Constitues eus principes V. Pro 10. Gr Constitues eus principes V. Pro

LA 141 LA 150 LA 135 LA 135

SGs 123 SGs 123 E 295 E 316

BV 40 BV 40 BV 40 fol. 119 BV 40 fol. 94 fol. 94 fol. 144

in ve - ri - ta - te quem ti - me - bo Pro pa - ti - bi fi - li

5. Co Pater cum essem 6. Co Tu puer 11. Gr Iustus non conturbabitur 12. Gr Propter veritatem

LA 140 LA (lacuna) LA 124 LA 132

E 252 E 269 E 284 SGs 126

BV 40 BV 40 fol. 73 fol. 89 BV 40 BV 40 fol. 51v fol. 104v

al - le - lu - ia vi - as e - ius -ba - bi - tur xri - [sto]

508 509 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Appendix 3a. Climacus with an s-shaped stroke as the uppermost note Appendix 3b. Climacus with an s-shaped stroke as the uppermost note (examples from the Gregorian repertory) (examples from the Beneventan repertory)

1. Gr Mittat tibi dominus 2. Al Iustus ut Palma 1. Communio Dixerunt (BV 40, fol. 139) 2. Offertorium Circuierunt (BV 40, fol. 122)

LA 147 LA 171

SGs 152

pa - ter de - se - ris et pel - li - bus ca - pri -

BV 40 BV 40 fol. 153v fol. 87v 3. Graduale Ecce magnum (BV 40, fol. 138v) 4. Offertorium Miraculo de tam (BV 40, fol. 125) au - xi - li - um [alleluia…]

3. Al Omnes gentes 4. Co Vovete

LA 174 LA 157 de - dit pau - pe - ri - bus - me - o ful - go - re

SGs 82 E 328 5. Antiphona Cum recubuisset (BV 40, fol. 6) 6. Communio O quantus (BV 40, fol. 139)

BV 40 BV 40 fol. 66v fol. 147v

[alleluia…] qui au - fert spi - ri - - lis suis ter - ge - bat qui - a pi - um est

5. Gr Timete 6. Al Deus iudex 7. Antiphona/ Communio Hodie exultat 8. Offertorium Paraclitus autem (BV 40, fol. 134v) (BV 40, fol. 71v) SGs 128 LA 173

BV 40 BV 40 fol. 73 fol. 158v do - mi - nus o - mni - bus - tet in no - mi - ne

do - mi - num [alleluia…]

510 511 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Appendix 4. The uppermost note of climacus is written with two parallel oblique strokes [ ]

1. Gr Convertere V. Domine 2. Of In omnem terram 7. Gr qui operatus V. Gratia dei 8. Al In exitu israhel

LA 151 LA 146 LA 136 LA (lacuna)

SGs 140 E 278 SGs 124 E 350

BV 40 BV 40 BV 40 BV 40 fol. 144v fol. 128 fol. 100 fol. 164v

Do - mi - ne e - o - rum fu - it [alleluia]

3. Al Dilexi quoniam 4. Of Protege domine 9. Al Quoniam deus 10. Al Venite exultemus

E 350 E 375 LA 175 LA 175

BV 40 BV 40 SGs 148 SGs 148 fol. 164v/165 fol. 55

BV 40 BV 40 fol. 163 fol. 163 Do - mi - nus Sa - cri - fi - [cium] [omnem] De - o

5. Of Oravi deum meum 6. Al In te domine speravi 11. Al Tu es petrus 12. Al Exultate deo 13. Al Te decet ymnus

E 327 LA 174 LA 170 LA 175 LA 174

SGs 146 SGs 123 SGs 147 SGs 147 BV 40 fol. 147

BV 40 BV 40 BV 40 BV 40 fol. 164v fol. 92 fol. 162 fol. 161v

Po - pu - li spe - ra - vi [me] - am De - o [sy] - on

512 513 CANTUS PLANUS 2014 PEATTIE — THE CLIMACUS IN BENEVENTAN NOTATION

Appendix 5. Elements of directionality and rhytmic nuance in the notation of the climacus in BV 40 Bibliography

Agustoni Luigi - Göschl Johannes B. (1987-1992), Einführung in die Interpretation des Gregorianischen graphic directionality frequent equivalents nuance Chorals, 2 vols., Regensburg: ConBrio. in E, SGs, LA

equal to the an important note; Albarosa Nino - Turco Alberto eds. (1991), Benevento, Biblioteca Capitolare 40. Graduale, Padova: La Linea 1. preceding note weighted or broadened (Codices gregoriani 1).

equal to the simple symbol, uppermost note Boe John (1983), ‘The Beneventan Apostrophus in South Italian Notation A. D. 1000-1100’, Early Music 2. preceding note is not weighted; this figure History 3, pp. 43-66. is used very frequently in the Gregorian repertory, but is mostly – (1992), ‘Review of The Beneventan Chant and Benevento 40: Graduale’, Notes 49, p. 519. higher than absent from the Beneventan 3. the preceding note Cardine Eugène (1982), Gregorian semiology, Sable-sur-Sarthe: Solesmes; or. ed. Semiologia gregoriana, 1968, Roma: Pontificio istituto di musica sacra. higher than an important note; 4. the preceding note weighted, broadened Le Codex VI 34 de la bibliothèque capitulaire de Bénévent, (XIe - XIIe siècle), Graduel de Benevent avec prosaire et tropaire (1937), Solesmes, Abbaye Saint Pierre (PalMus 15). lower than no consistent pattern of use; 5. the preceding note meaning appears to be strictly Fischer Rupert (1991), Die rhythmische Aussage von Benevento 40 in Albarosa e Turco 1991, pp. ix-xii. directional Graduale Triplex (1979), Solesmes: Abbaye Saint-Pierre.

Haug Andreas (2020), Towards a semiotically informed transcription practice in James Borders ed., IMS lower than light upper note; frequently Study Group Cantus Planus, Papers read at the xvii meeting, Venice, Italy, 28 July - 1 August 2014, 6. the preceding note* employed on unaccented syllables Venezia, Fondazione Levi, pp. 137-142.

equal (or higher) an important note; Peattie Matthew (2014), Constructing a Typeface of Medieval Neumatic Notation: Challenges of Representation than preceding note weighted, broadened, weight and Translation in Armin Karim - Barbara Swanson eds., Chant and culture. Proceedings of the 7. towards a lower note? ** conference of the Gregorian Institute of Canada, Lions Bay: Institute of Mediaeval Music, pp. 245-262.

higher or lower very infrequent in Gregorian ligation of two notes at the unison; 8. than preceding note repertory; pervasive in south a doubling of length; Italian repertories perhaps rearticulated (?)

* In V-CVbav Ottob. lat. 145, this figure is used occasionally in unison with the previous note (see example 8 above). + This form is used exclusively in compound. ** The uppermost graphic of no. 6 and 7 are also used in isolation and ligation.

514 515 Bibiana Gattozzi

The hymn in Beneventan manuscripts

Even though in the earliest years of the church, contention often surrounded the liturgical use of hymns – strophic, metrical Latin poetry with texts not drawn directly from sacred scripture – it was only a short time before hymns were used in the divine office at each canonical hour (Boynton 2005, pp. 19-21). Despite its liturgical importance and presence in manuscripts preserving chants, texts, and rubrics for the divine office, a thorough survey of this genre in medieval southern Italy through an examination of the main sources for the hymns, a cataloguing and transcription of melodies, and repertorial and melodic comparison is still far from complete. The study of regional hymn repertoires, especially those of geographical areas in which local, non-Gregorian liturgies played a prominent role, is not new (Gutiérrez 2004, pp. 216-243). This fact, coupled with the variety of recent scholarship on the hagiographical, paleographical, artistic, liturgical, and political contexts of southern Italy makes a similar survey of the hymn in this milieu particularly germane.1 Indeed, this type of repertorial survey is essential for addressing the following questions: what were the liturgical, theological, and political functions of the hymn in southern Italy when the earliest hymn collections were copied in this region? How was hymn repertoire transmitted in southern Italy? Was this genre affected, as other plainchant genres of the local, Old Beneventan liturgy, by liturgical reforms? What are the characteristics of unique or local hymns and to what extent are they present in manuscript sources? How do local hymn texts and/or melodies articulate local devotional and/or musical traditions?

As a step towards a more systematic repertorial survey of hymns in Beneventan manuscripts, I propose to identify the main southern Italian sources for hymns, showing how the repertoire of hymns they contain is, to some extent, dictated by manuscript typology. I will present the manuscript Benevento, Biblioteca Capitolare I-BV (henceforth BV) 37, one of the earliest sources for hymn texts and melodies from Benevento, as a case study through which to categorize hymn texts as widespread, local, or unique. This categorization, while not new in studies of chant, has nevertheless not been applied systematically to the hymn repertoire in Beneventan sources despite its usefulness for reconstructing the history, function, and transmission patterns of liturgical genres. I will compare the melodies of local

1. A few important recent publications include Hamilton (2010), Oldfield (2014), Ramseyer (2006), and Zschomelidse (2013).

517 CANTUS PLANUS 2014 GATTOZZI — THE HYMN IN BENEVENTAN MANUSCRIPTS hymns shared by BV 37 and another important hymn source, BV 42. Both manuscripts Table 1. Main manuscript sources for hymns in Benevento, Montecassino, and central and southern Italy, eleventh-thirteenth centuries were destined for female monasteries in Benevento and probably copied at Santa

Sofia, although BV 42 was written a century later (Mallet and Thibaut 1997, 2, pp. Manuscript Type Origin Destination Date 193-235, 255-280). The same hymn texts that have different melodies in each of From Benevento (= BV) these sources raise the question as to whether liturgical reforms or devotional and V-CVbav Ott.lat. 145 Composite Santa Sofia, BV MC? BV, Santa eleventh century ca. musical considerations prompted changes through time in the melodic assignments Sofia? first-second half of local hymn texts. This study draws on the work of collecting and editing hymn I-BV 37 Composite Santa Sofia, BV BV, female eleventh century ca. monastery second half texts begun by Bruno Stäblein in Monumenta monodica medii aevi (Stäblein 1956, I-Nn VI G 312 Composite Santa Sofia, BV BV eleventh century ca. 1), by Blume, Dreves, and Bannister in Analecta hymnica (Blume and Dreves 1888- last quarter-thirteenth 1909), on the work of cataloguing hymn sources by region conducted by James century I-Nn VI E 43 Composite* Santa Sofia, BV BV Santa Maria 1099-1118 Mearns in his study of early Latin hymnaries (Mearns 1913), as well as on the more recent and ongoing survey of the Italian hymn by Giacomo Baroffio (Baroffio, n.d.). V-CVbav Vat.lat. 4928 Composite* Santa Sofia, BV BV twelfth century ca. Monumenta monodica medii aevi separates the main hymn repertoire of Western I-BV 42 Composite Santa Sofia, BV BV, San Pietro twelfth century Europe by region; it references, among its sources, the twelfth-century hymnary of intra muros second half Gaeta in Beneventan script: Rome, Biblioteca Casanatense, 1574. It also cites hymns From Montecassino (= MC) I-MC ms. 506 Hymnary MC MC ~1050 from other southern Italian manuscripts. Similarly, Dreves and Blume cited thirteen manuscripts from central and southern Italy in Analecta hymnica for their editions I-Nn VI F 2 Composite MC MC; Naples3 1090s of hymn texts. However, many sources in Beneventan script containing important F-Pm 364 Composite* MC MC 1099-1105 collections of hymns, such as BV 37 and BV 42, have yet to be studied in detail. Before I-MC ms. 559 Psalter-Hym. MC MC eleventh-twelfth century ca. looking at these two important Beneventan hymn sources specifically, it will be V-CVbav Urb.lat. 585 Composite* MC MC 1099-1105 necessary to discuss the types of sources that transmit hymns in the Beneventan US-MAL Ms. Ludwig IX 1; Composite* MC MC 1153 zone, as source type influenced the choice of hymns included in the manuscript. 83.ML 97 A-Wn 1106 Composite MC MC twelfth century ca. Hymns in southern Italy are found in sources for the divine office such as the From places other than Benevento and Montecassino manuscripts BV 19 and BV 20, a combined missal-breviary; in antiphoners such CH-Zz Rh. 913 Composite Subiaco Subiaco eleventh century as BV 21 and Biblioteca dell’Abbazia di Montecassino (= MC) 542; in breviaries; in first quarter composite sources, some of which contained ordinals, such as Biblioteca Apostolica V-CVbav Vat.lat. 3797 Hymnary Fonte Avellana Fonte Avellana eleventh century

Vaticana (= V-CVbav) Vat. lat. 4928; in composite office sources without ordinals, such CH-Zz Rh. 82 Composite Farfa Farfa eleventh century as BV 37; and in hymnaries such as MC 506. Hymns are also sometimes found, usually F-Pn 1092 Hymnary Narni Narni eleventh century first half only cited and in limited numbers, in books not exclusively for the divine office, such as graduals. They are also found in sources with more circumscribed purposes such V-CVbav Vat.lat. 7172 Composite Narni Narni eleventh century first half as psalters and processionals. The primary Beneventan-script sources for hymns V-CVbav Chig.C.VI 177 Composite Farfa Farfa, Subiaco 1050-1060 between the eleventh and thirteenth centuries are listed in table 1, along with some I-MC ms. 420 Hymnary Subiaco Subiaco 1068-1120 of the main hymn sources from central Italy because these often shared local hymn I-FRa A.209 Composite Farfa Farfa twelfth century last quarter repertoire with Beneventan hymnaries. This list is not exhaustive; the sources here I-Rn 175 (Farfa 4) Composite Farfa Farfa ~1100 all contain hymns whose texts are copied out in full. Breviaries, antiphoners, and graduals are not included, even though these cite hymns. Composite sources with an I-Rc 1574 Hymnary Gaeta Gaeta twelfth century asterisk indicate that the manuscript also contains an ordinal. I-BAsn 15 (A 13) Hymnary Bari Bari thirteenth century

I-Nn VI G 29 Hymnary Naples Naples thirteenth century

2. Folios 1-46 are part of an eleventh-century lectionary/collector and hymnary, while folios 47-59 are part of a rituale from the thirteenth century. 3. Newton 1999, pp. 321-322.

518 519 CANTUS PLANUS 2014 GATTOZZI — THE HYMN IN BENEVENTAN MANUSCRIPTS

Notable in the main surviving sources for hymns used at Montecassino is the early complete hymnaries such as this were widespread in the Beneventan region and date of most of the sources compared to the ones from Benevento (except BV 37 and later dismembered or rearranged into libelli according to the needs of the religious V-CVbav Ott. lat.145) and the high occurrence of composite office sources that include community to which they belonged, or to resemble other composite office sources. ordinals. The ordinal manuscripts reflect the efforts of abbots at Montecassino to Let us explore this suggestion further. The hymnary in BV 37 was obviously taken organize and unify liturgical practice. Sources such as these could have been used from a different source: the first part of BV 37, folios 1-18v, was copied by a late as models for other institutions (Kelly 2008, pp. 49-50). Indeed, the hymn collections eleventh-century scribe, while the hymnary on folios 19-54v was copied by a in the manuscripts with ordinals tend to be very standardized in terms of order and different, eleventh-century scribe (with a few twelfth-century replacements, folios contents, and contain fewer local or unique hymns compared to other hymnaries 47-52v), and the final section of the manuscript containing litanies, a martyrology- from central and southern Italy. Other prominent hymn sources in Beneventan script necrology, and the Rule of Saint Benedict was copied by two hands from the end or notation come from Naples, Narni, and Farfa. There are also other fragmentary of the eleventh century. As a composite source, BV 37 was compiled from different sources that transmit hymns in Beneventan script not listed here. I have taken the time gatherings containing different types and genres of texts, rubrics, and chants useful to summarize these Beneventan-script hymn sources to emphasize the fact that only for a monastic liturgy. The items in BV 37 highlight the liturgical and devotional by cataloguing the repertoire of hymn texts in each of these manuscripts, comparing priorities of the monastery for which it was destined, San Pietro intra muros. The these catalogues, and searching for concordances in sources from other regions, can incomplete hymn collection of BV 37, when ordered according to the liturgical we identify and categorize hymns as widespread, local, or unique. If the unique hymn year, includes hymns from the feast of the Holy Innocents to that of Saints Peter texts have notated melodies, these melodies can be examined to see whether, at times, and Paul. The hymn ordering in BV 37, originally according to liturgical calendar, different melodies were used in southern Italy for more widely distributed hymns; or, was significantly changed due to subsequent reordering of the folios and gatherings. if they are unique, whether they might be examples of local musical style. With this Table 2 shows the gatherings with the hymns in liturgical calendar order. in mind, we may now zoom into the source BV 37 and use it as a case study to show how this task may be approached, as well as to highlight some unusual features of the Table 2. Hymn contents and gathering structure of BV 37 manuscript in comparison to other hymn sources, and to examine melodic differences between hymns in BV 37 and BV 42. fol. 47-fol. 52v fol. 36r-v fol. 35r-v fol. 39-fol.46v fol. 20- fol. 34r-v fol. 19r-v fol. 53-fol. 58v fol. 26v fol. 27-fol. 32v fol. 37r-v fol. 38v fol. 33r-v In manuscripts, hymns can be found cited as rubrics with only a textual incipit, as often happens in breviaries such as BV 19 and BV 20, or as fully copied texts without notation, as in I-Nn VI E 43 (Boynton 2003, pp. 115-135). Hymns are often found with their first strophe copied with musical notation and the rest of the text copied below without musical notation, as in BV 42. Of all these sources, BV 37 is one of the earliest Holy Innocents Vincent Gregory (end) Saint Benedict Easter Eustasius (end) and the one in which hymns are copied most completely, namely with full musical Sylvester Scholastica Septuagesima Mark Vitus (begins Epiphany Sabinus Nazarius Annunciation Ascension fol. 34, ends fol. 27) notation for each strophe of text. This layout is unusual for hymns, whose strophic Maurus (begins on fol. 19, Sunday Lent Passion Sunday Pentecost Bartholomew melodic structure does not necessitate the re-copying of the melody for every single Sebastian ends on fol. 37) Ferial Lent Palm Sunday John Baptist Agnes Gregory (end on fol. 38v) Maundy Thursday John and Paul strophe. One advantage of this layout is that the melodies follow more closely the (beginning) Vincent Benedict Easter (part) Peter and Paul nuances of the text, since liquescent neumes, for example, could be indicated on (beginning) (beginning) appropriate syllables as the text required. This could also serve didactic purposes. Since these nuance neumes were much more common early on and gradually fell out of use, the format of the hymns in BV 37 might reflect the copying priorities of The confusing folio numbering at the top, which reflects the foliation given after eleventh-century scribes who wanted to indicate these melodic signs meticulously the reordering of the gatherings, shows the extent of the rearrangement of folios. (ibidem, pp. 119-124). In addition to these reasons for fully copying hymns, I would The gathering with folios 47-52 was a twelfth-century addition to the collection; it like to suggest another one: the hymnary portion of BV 37, a composite source was probably at this time that the codex was put together and reordered. All hymns made of different smaller collections of liturgical items, may have been taken from except those on this replacement gathering and the hymn for the feast of Saint another source that exclusively contained hymns. Such a single-genre source would Nazarius, which was added interlinearly later, are notated. There is a lacuna of five justify the layout of the hymns with complete musical notation. Perhaps at one time folios after folio 36 (feast of Saint Vincent), one folio after folio 21 (Saint Mark), and

520 521 CANTUS PLANUS 2014 GATTOZZI — THE HYMN IN BENEVENTAN MANUSCRIPTS five folios after folio 26 (Pentecost), which accounts for the omission in BV 37 of some asterisks are hymns that have regional concordances, but only with manuscripts that important feasts such as Saint Barbatus and the dedication of Saint Michael, feasts were copied and/or used outside of Benevento itself, e.g., from Montecassino and/or which have hymn assignments in the more complete collection in BV 42. Another central Italy (exceptions are noted in parentheses). The cross indicates the widespread indication that BV 37 was once a complete and separate hymnary is provided by its presence of the hymn even in sources outside of Italy. The hymns that have a different hymn contents. For some feasts in BV 37, alternate hymns are included in addition to assignment in BV 37 than in other manuscripts are indicated. Those with neither cross the ones listed as the norm for that feast. These are summarized in table 3. nor asterisk are hymns that are known elsewhere, but only in the regions of central and southern Italy. Due to the high number of local texts in these alternate hymns Table 3. Alternate hymns for certain feasts in BV 37, some with only local and/or regional concordances (indeed, there is at least one local hymn for all these feasts except those of Easter, Pentecost, and Saints Peter and Paul), we see that alternate hymns for feasts of local Feast No. of hymns Hymns importance maintained local flavor in the divine office at Benevento, even though the Holy Innocents 3 Salvete flores martyrum+ divine office was becoming more standardized in southern Italy. Verbi patri principium* Infantum diem martirum* (I-Nn VI E 43) Why, in BV 37 as in most other Beneventan-script hymn sources, is it useful to identify Saint Vincent 3 Adest miranda* (V-CVbav Ott.lat. 145) hymns that are part of a ‘core repertoire’ widespread throughout Europe, hymns that Agonitheta nobilis (different assignment) Beatus vir Vincentius* circulated primarily in Beneventan manuscripts, and hymns that are unique to a given manuscript? Consideration of these three different repertorial distribution categories Lent, matins 2 Iam lucis splendor rutilat Iam Christus sol+* (I-Nn VI E 43) reveals the priorities governing the selection, adoption, adaptation, and composition of hymns before their compilation into composite manuscript sources. Moreover, a Easter, vespers 3 Ad cenam agni provide+ Rex sempiterne domine+ (different assignmnent) melodic investigation of unique or local hymns provides a glimpse at the musical style Iesu nostra redemptio+ or compositional priorities of local religious communities. Usually the hymns that Ascension 4 Optatus votis omnium+ form part of a more widely-distributed repertoire are those that have ‘ordinary’ or fixed Iam Christus ascendit polum assignments for given canonical hours and/or feasts. They are not usually assigned to + Festum nunc celebre special feasts or feasts of saints with local importance and are shared amongst many Funeris victor triduo* (V-CVbav Ott.lat. 145) sources both within and outside the Beneventan circuit without great differences in + Pentecost 3 Veni creator Spiritus text and melody. Examples of widespread hymns in BV 37 are, unsurprisingly, mostly Beata nobis gaudia+ Iam Christus astra ascenderat+ from the temporal cycle, such as those for the Epiphany, Lent, Passion Sunday, Palm Sunday, Easter, the Ascension, and Pentecost. The sanctoral feasts with widespread Saint Bartholomew 2 Ad laudem Christi procerum Gaudium mundi Christe hymns are those for saints venerated universally, such as Saints Sebastian, Agnes, Benedict, the Annunciation, Saints Michael, Scholastica, John the Baptist, and Saints Saint John Baptist 3 Ut queant laxis+ Almi prophete progenies+ Peter and Paul. The hymns in BV 37 that have limited or no presence in manuscripts Deus qui verbi angelo (unique) outside the Beneventan zone before the fourteenth century are those for feasts of saints Saints Peter and Paul 3 Aurea luce et decore roseo+ with great local importance. Sanctoral and temporal feasts in which multiple hymns Doctor egregie+ (division of Aurea luce) were used (such as the Ascension) also allowed for the inclusion of local hymns besides + Apostolorum passio diem the more widespread repertoire; the local character of certain alternate hymns has

*Asterisk denotes hymns with regional concordances, but only in manuscripts copied outside Benevento already been mentioned. Indeed, from the concordances of the local hymns assigned (e.g., Montecassino or central Italy). to specific saints’ feast days in BV 37, the hymns for Saint Bartholomew (Ad laudem The sign + denotes widespread presence of the hymn even in sources copied outside Italy. Christi procerum and Gaudium mundi Christe) are represented in the greatest number of Beneventan sources; these hymns are present also in the two hymnaries from Often, these alternate hymns are local to the Beneventan zone. The hymnaries in Narni, and Ad laudem Christi procerum is present in I-Nn VI G 29. Saint Bartholomew, composite volumes that were planned as composite volumes at the time of their whose relics were contested between Benevento and Naples, was well-known and copying usually contain fewer local hymn texts. One of the three alternate hymns for much venerated in Benevento (Oldfield 2014, pp. 21-23, 72). The hymn O clara mundi the feast of Saint John the Baptist, Deus qui verbi angelo, is unique to BV 37. Those with lumina for Saints John and Paul is only present in BV 21 and BV 37, while that for Saint

522 523 CANTUS PLANUS 2014 GATTOZZI — THE HYMN IN BENEVENTAN MANUSCRIPTS

Eustasius, Eustasii magni martyris, is only present in BV 37 and BV 42. Saint Eustasius melody of Agonitheta nobilis is likewise a common one, but not the same as the one in BV might have been venerated in a special way at female monasteries in Benevento, 37. This melody is used for many other hymns including that for Saint Sylvester, Sylvestri wherefore the hymn is present in the manuscripts destined for these monasteries and almi presulis, a hymn with regional distribution in central and southern Italy. The use preserved through time in hymn collections. of different melodies for the same hymn texts was very common throughout Europe. Nevertheless, although both melodies were probably borrowed for use with this local The chant Agonitheta nobilis presents an interesting case of local adaptation of hymn hymn text, the change of melodies from BV 37 to BV 42 shows the transformation and repertoire, and can serve as an example of the issues of transmission involving the adaptation undergone by local hymn texts which were recopied in succeeding centuries hymn repertoire in the Beneventan zone. Agonitheta nobilis is present in BV 37, BV to fulfill changed local liturgical needs. We cannot necessarily say that the melody in 42, and V-CVbav Ott. lat. 145 from Benevento, in the hymnaries from Narni, in CH-Zz BV 37 is the one originally assigned to this text; it could be that scribes, knowing of the Rheinau 91, and in the Beneventan ordinal source V-CVbav Vat. lat. 4928. However, assignment of the text to the feast of Saint Cyriacus in sources such as V-CVbav Vat. lat. only BV 37 assigns this hymn text to the feast of Saint Vincent, whereas the others 4928 thought it best to use a different melody rather than the one that had been used for assign it to the feast of Saint Cyriacus (except V-CVbav Ott. lat. 145 which assigns it to Saint Vincent previously in Benevento. However, it could also be that the scribe of BV the common of one martyr). Moreover, in BV 37 the melody of this hymn is different 37, adopting the text for Saint Cyriacus (or for the common of one martyr) for the feast from that in BV 42, as shown in examples 1 and 2. of Saint Vincent, thought to assign it a different melody. More comparative research is needed to establish the directionality of such melodic transformations, but instances Example 1. Agonitheta nobilis in BV 37, folio 36r-v for the feast of Saint Vincent such as these of differences between melodies assigned to the same texts in manuscripts from different geographic locations or centuries gives clues about the nature of melodic and textual transmission of the hymn repertoire in the Beneventan zone.

Before examining another instance of a hymn with a different melodic assignment but with the same text in BV 37 and BV 42, it is interesting to note that some of the local hymn repertoire in BV 37, although not found in other manuscripts from Benevento proper, are instead found in sources from Montecassino, Naples and central Italy, as shown in table 4, p. 526 (Infantum diem martyrum is an exception). This correlation has heretofore not been examined as a phenomenon per se. Usually, the concordances are between BV 37 and the two almost identical manuscripts from Narni, V-CVbav Vat. Example 2. Agonitheta nobilis in BV 42, fol.151 for the feast of Saint Cyriacus lat. 7172 and F-Pn 1092, and with the early source V-CVbav Ott. lat. 145. These hymns are assigned for feasts of the sanctoral (excepting Rex sempiterne domine for Easter and Funeris victor triduo for the Ascension), and more specifically for feasts of saints with strong local cults. Again among these we see two hymns for Saint Vincent, as well as one for Saint Maur who was venerated in many Benedictine abbeys including Montecassino. Saint Vitus was especially important in the area of southern Italy known as Lucania, his native land, in present-day Basilicata.

From these concordances we can draw two different scenarios for the transmission of hymns in the Beneventan zone: first, at the time that BV 37 was copied, the hymn repertoire of southern Italy was more expanded, with many instances of shared The melody of Agonitheta nobilis in BV 37 resembles that of many hymns of widespread repertoire between sources within and without Benevento, but that by the time BV 42 distribution; in fact it closely resembles hymns such as Pange lingua in melodic outline. was copied, many local saints’ hymns were omitted in favor of greater uniformity. The The well-known hymn melody was probably borrowed and adapted to the new text in presence of the hymn Infantum diem martyrum for the feast of Holy Innocents in only BV 37, which was then assigned for the feast of Saint Vincent, whose local importance one other Beneventan source, I-Nn VI E 43, copied in the same period as BV 37, may in southern Italy and at this monastery necessitated a hymn in his honor. In BV 42, the support this hypothesis. Infantum diem martyrum probably was transmitted to Naples

524 525 CANTUS PLANUS 2014 GATTOZZI — THE HYMN IN BENEVENTAN MANUSCRIPTS and Benevento from Montecassino, given its presence also in early hymnaries such as The second possibility is that these local hymns were gathered into BV 37 from outside MC 506 and V-CVbav Urb. lat. 585. Perhaps sources such as the ordinal V-CVbav Vat. of Benevento because – assuming that the hymns in BV 37 were once part of a more lat. 4928 were responsible for the disappearance of these hymns from the Beneventan complete, separate hymnary – BV 37 was meant to be as complete as possible for orbit, since these hymns appear neither in the Beneventan ordinal nor in later sources. didactic, preservation, or presentation purposes, something supported by the presence As a source that prescribes the chants for the divine office and their order, the ordinal of musical notation on every strophe of the hymns in this source. Thus, different must have influenced the type of repertoire of a specific genre that was included in hymns were borrowed even from outside Benevento, indeed from as far as Spain, subsequent manuscript collections. as the presence of the hymn for Saint Vincent, Adest miranda passio levita, in the eleventh-century hymnaries GB-Lbl 30851 and E-Mn 10001 attests. But there is no Table 4. Concordances between hymns in BV 37 and sources from outside Benevento surprise here; since Saint Vincent himself came from Spain, why not include a hymn from his native land? The hymn Festum beati martyris, like Agonitheta nobilis, has Title Feast Concordances been included in table 2 even though properly speaking it is also shared with another Infantum diem martyrum Holy Innocents V-CVbav Vat.lat. 7172, Beneventan manuscript, BV 42. One would therefore think that the melodies for this F-Pn 1092 hymn in both manuscripts are the same; given its absence from many codices in V-CVbav Urb.lat. 585 Benevento proper, we might be tempted to assume that the hymn was copied from BV I-MC ms. 506, I-Nn VI R 43 37 or a similar, now-lost exemplar by the scribes of BV 42 a century later. However, Verbum patris principium Holy Innocents V-CVbav Vat.lat. 7172, like Agonitheta nobilis, Festum beati martyris is an example of a hymn with the same F-Pn 1092, CH-Zz Rh. 91 text but different melodies in BV 37 and BV 42, as shown in examples 3 and 4. Adest celebritas nobis Saint Maur V-CVbav Vat.lat. 7172, F-Pn 1092 Example 3. Festum beati martyris in BV 37, fol. 21r-v for the feast of Saint Mark V-CVbav Ott.lat. 145

Adest miranda passio Saint Vincent V-CVbav Vat.lat. 7172, F-Pn 1092 V-CVbav Ott.lat. 145 GB-Lbl 30851, E-Mn 10001

Beatus vir vincentius Saint Vincent V-CVbav Vat.lat. 7172 F-Pn 1092

Claret sacrata iam dies Saint Gregory of Spoleto V-CVbav Vat.lat. 7172 F-Pn 1092

Angusta vite tempora Saint Nazarius V-CVbav Vat.lat. 7172 F-Pn 1092 Example 4. Festum beati martyris in BV 42, folios 136v-137 for the feast of Saint Mark Rex sempiterne domine Easter V-CVbav Vat.lat. 7172 F-Pn 1092, CH Z-z Rh. 82, I-Nn VI G 29

Funeris victor triduo Ascension V-CVbav Vat.lat. 7172, F-Pn 1092 V-CVbav Ott.lat. 145

Festum beati martyris Saint Mark I-BV 42, V-CVbav Vat.lat. 7172 F-Pn 1092, I-Nn VI G 29 CH-Zz Rh.91

Alma beati martyris viti Saint Vitus V-CVbav Vat.lat. 7172 F-Pn 1092 The melody in BV 42 resembles that of the hymn Verbum salutis omnium, while that V-CVbav Ott.lat. 145 from BV 37 is not concordant with any other hymn melodies that I have observed.

526 527 CANTUS PLANUS 2014 GATTOZZI — THE HYMN IN BENEVENTAN MANUSCRIPTS

The melody in BV 37 might have been a local composition that was then replaced Bibliography by a different melody in later manuscripts. This further supports the view that the hymn repertoire in the Beneventan zone changed through time to adapt local texts Baroffio Giacomo (n.d.), Corpus Hymnorum Italicum, http://www.cantusgregorianus.it/vecchiosito/ to new melodies, possibly in an effort to unify liturgical practice by eliminating local strumentiricerca/C_H_I_-_Corpus_Hymnorum_Italicum_inventario.pdf (17 February 2015). compositions in favor of more well-known melodies. Boynton Susan (2003), ‘Orality, Literacy, and the Earl Notation of the Office Hymns’,J ams 56, pp. 99-168.

– (2005), The Theological Role of Office Hymns in a Ninth-Century Trinitarian Controversy in Benoît- Besides the hymns that are shared between BV 37 and other sources, there are a few Michel Tock ed., In principio erat verbum. Mélanges P. Tombeur, Turnhout: Brepols (Textes et Etudes unique hymn texts in BV 37: Presulem sanctum reboate for Saint Sabinus and Deus qui du Moyen Âge 25), pp. 19-44. verbi angelo for Saint John the Baptist. Another hymn, Decessit ut mitissimus, is merely Dreves Guido Maria - Blume Clemens - Bannister Henry Marriott (1886-1922), Analecta Hymnica, 55 vols., a division of the hymn O clara mundi lumina for Saint John and Paul which is only Leipzig: Reisland. present in BV 21 and BV 37. The hymn Presulem sanctum reboate for Saint Sabinus Gutiérrez Carmen Julia (2004), The Hymnodic Tradition in Spain in Andreas Haug - Christopher März (example 5) has the same melody of the hymn Confessor Domini Maure, a hymn that is - Lorenz Welker eds., Mmmae, 4: Der lateinische Hymnus im Mittelalter, Kassel: Bärenreiter, also local to southern and central Italy. The melody is common to many antiphons and pp. 216‑243. alleluias, another example of how local hymn texts were adapted to existing melodies. Hamilton Louis I. (2010), A Sacred City: Consecrating Churches and Reforming Society in Eleventh-Century Italy, Ithaca: Cornell University Press. Example 5. Presulem sanctum reboate melody in BV 37, fol. 19v and 37 for the feast of Saint Sabinus Kelly Thomas Forrest (2008) The Ordinal of Montecassino and Benevento: Breviarium sive ordo officiorum, 11th century, Fribourg: Academic Press (Spicilegium Friburgense: Texts Concerning the History of Christian Life, 45).

Mearns James (1913), Early Latin Hymnaries, Cambridge: Cambridge University Press.

Oldfield Paul (2014), Sanctity and Pilgrimage in Medieval Southern Italy, 1000-1200, Cambridge: Cambridge University Press.

Ramseyer Valerie (2006), The Transformation of a Religious Landscape: Medieval Southern Italy, 850-1150, By taking BV 37 as a case study, observing its repertoire of hymns and comparing it to Ithaca: Cornell University Press. other hymn sources from the Beneventan zone, we can conclude that, besides hymns Stäblein Bruno ed. (1956), Mmmae, 1.1: Die Hymnen, Kassel: Bärenreiter. of widespread distribution present in the liturgy of the divine office in southern Italy, there were a significant number of local hymn texts and melodies. Some of these were Zschomelidse Nino (2013), Art, Ritual, and Civic Identity in Medieval Southern Italy, University Park: Pennsylvania State University Press. specific to single religious communities and most often were associated with feasts of saints with strong local cults. Sometimes the same local hymn texts persisted through time but were adapted to different melodies. Yet, as the divine office became more standardized in manuscripts such as those with ordinals, a number of local hymns and regional hymns that had been shared between Benevento and other areas of Italy were excised from hymnaries. This might have prompted changes in the format of hymnaries, which once might have looked like the fully-notated BV 37. Nevertheless, by tracing the presence and melodic characteristics of hymn texts and the copying layout in extant manuscripts, it is evident that some local hymns had great prestige in the Beneventan zone and enjoyed longevity due to the flexible re-adaptation of texts and melodies according to liturgical and devotional needs.

528 529 CANTUS PLANUS 2014

IX Byzantium and the East

530 531 Svetlana Kujumdzieva

The tropologion Vaticanus Graecus 771

The tropologion is considered the earliest known extant chant book that has preserved layers of Jerusalem hymnography and liturgy from the fifth-sixth century and was in use until about the twelfth century.1 The study of the tropologion is still in its infancy. Thirteen manuscripts in Greek of this type of book have become known quite recently. They were written between the eighth-ninth and the twelfth century. The manuscripts bare the title «Tropologion» either in their initial rubrics (figure 1) or in their inscriptions left by the copyists.

Most of them are preserved at the library of Saint Catherine’s Monastery on Mount Sinai. Two of them are from the Vatican Apostolic Library in Rome (V-CVbav), Vat. gr. 771, and Vat. gr. 2008.2 Both are with notation. Here I am going to present the former, Vat. gr. 771. I have discussed the latter in a contribution for the Festschrift dedicated to Prof. Gerda Wolfram (Kujumdzieva 2011, pp. 187-205). In all probability Vat. gr. 771 was written by the end of the eleventh century by one of the abbots of the Grottaferrata monastery (Bertonière (1997, p. 95 and 158): it displays characteristics that are typical for the scriptorium there. Stefano Parenti (2000, pp. 543-556) links the manuscript to Grottaferrata monastery as well. According to him it was written by the abbot Nilus the Second, who died in 1135. It is well known that the Grottaferrata monastery retained its Byzantine rite and traditions that were brought by the founder of the monastery, Saint Nilus, by the very beginning of the eleventh century. The abbot Nilus the Second left two inscriptions in the Vatican manuscript. In both of them he calls the book «tropologion». The first inscription (figure 2, p. 534) is on the bottom of fol. 118v. It refers to the great kanon by Andrew of Crete. We read: «Ζήτει εἰς τὴν ἀρχὴν τοῦ τροπολογίου», that is: «Look for it in the beginning of the tropologion». The second inscription (figure 3, p. 535) is on the bottom of fol. 288v. It refers to the photagogikon, which is performed after the ninth ode of the kanon. Nilus writes: «Ζήτει τὸ φωταγωγικὸν εἰς τὸ τέλος τοῦ τροπολογίου», that is: «Look for the photagogikon at the end of the tropologion».3

1. There is already a rich literature about this book, see – selectively – Metreveli, Čankieva and Hevsuriani (1980); Jeffery (1991, pp. 52-75; 2001, pp. 147-211); Renoux (2001, p. 311); Frøyshov (2013); Kujumdzieva (2013, pp. 118-129). See also the cited literature there.

2. I had a rare chance to work twice, in 2003-2004 and 2012, as a fellow with the Vatican manuscripts in microfilm, which are kept at The Vatican Film Library (Vfl) in the main University library, The Pius xii Memorial Library, in Saint Louis University in Saint Louis, Mo; about the manuscripts in Vfl see Ermatinger (1993). My deepest gratitude goes to the whole staff of the Vfl and especially to Dr. Gregory Pass and Dr. Susan L’Engle. I have worked with Vat.gr. 771 on microfilm at this library. About my work with the musical manuscripts at the Vfl see Kujumdzieva (2008, pp. 43-155).

Figure 1 3. The inscriptions are published by Momina and Тrunte (2004, p. 44) but on wrong folia. ET-MSsc gr. 607, ninth century, fol. 3 The rubric says that the manuscript is a tropologion for two months, March and April 533 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

Figure 2 Figure 3 Vat. gr. 771, fol. 118v Vat. gr. 771, fol. 288v The first inscription of Nilus on the bottom of the sheet The second inscription of Nilus on the bottom of the sheet

534 535 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

Figure 4 Figure 5 Vat. gr. 771, fol. 1, the beginning: stichera idiomela in theta-notation Vat. gr. 771, fol. 1v On the top there is the rubric for Sunday vespers Chants by Leo vii in theta-notation

536 537 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

The beginning of the manuscript is not preserved and we do not know how it was defined in its initial rubrics. According to Nilus’ inscriptions we could suggest that the manuscript was defined as a tropologion there as well. Vat.gr. 771 contains a repertory of the triodion-pentekostarion, that is, for the movable feasts of the year. It counts 298 folia and measures 24,7 × 17, 3 (Devresse 1950, pp. 286-287). The first rubric preserved (figure 4, p. 536), which one reads on the top of fol. 1, prescribes chants for Sunday vespers.

I identified the chants for the third week of Lent. Chants for weekdays from Monday through Friday for the fourth week of Lent follow. Most of them have a theta sign. All are stichera idiomela.4 The stichera in theta-notation are given for both weekdays and Sundays. The notated stichera follow up to Good Friday. This first part of manuscript ends on fol. 8. The sign theta is written at the end of the phrases. It is written either independently or with double bareia put right or left of the theta (figure 5, p. 537).

On fol. 8v a new large rubric introduces the beginning of Sunday of Prodigal Son and specifies that the latter is Κυριακὴ« πρὸ τῆς Ἀπόκρεω τοῦ Ἀσώτου», before Meetfare Sunday. This part of the manuscript, let say the second one, contains the repertory for the preparatory season of Lent. It starts with a kanon by the great hymnographer of the ninth century Joseph the Hymnographer. Chants for Meetfare and Cheesefare weeks follow. They are of the genres of kanons (all of them are triodia), stichera and kathismata. The kanons are attributed to Clement, Theodore the Studite and Joseph (figure 6, p. 538).

Many of the stichera and kathismata are attributed to Theodore alone. On fol. 32v a new rubric announces the beginning of Lent: «Τεσσαρακοστῆς τέσσαρα». It contains the same genres: kathismata, stichera and kanons. The kanons are both triodia and full with nine or eight odes. They are for the days from Monday through Saturday and are attributed to the same three authors: Clement, Theodore and Joseph. Kanons only are written down for Sundays. The first Sunday of Lent is designated as «first» and «Orthodox». Rest of the Lenten Sundays are counted from second to sixth without any other designation (such as Sunday of Saint Gregory Palamas, of the Cross, of Saint John, etc. as they are marked today). The Great kanon by Andrew of Crete, which the first Nilus’ inscription refers to, is in plagios second mode. It is indicated for the fifth Sunday in Lent. According to Gabriel Bertonière (1997, p. 89) this indication makes the Vatican manuscript unique because it appears to be among the earliest one that says where this kanon has to be performed; the incipit of the kanon is «Πόθεν ἄρξομαι». According to Bertonière, there is no unity in manuscripts in terms of the kanon for the fifth Sunday. Many of the stichera and kathismata for weekdays are attributed to

Figure 6 4. The notation is announced by Raasted (1962, p. 203). The manuscript is not included in the checklist of the Vatican Vat. gr. 771, fol. 69v musical manuscripts by Touliatos-Banker (1987). Bertonière (1997, p. 65) says that the idiomela stichera are grouped A combined kanon by Clement, Studite [Theodore] and Joseph together in the beginning of the manuscript but he does not mention the theta-notation.

538 539 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

Theodore alone again. Two kanons by the old Sabaite masters, Kosmas the Monk and and structures that were established by the Sabaite monks; the Studites edited their Andrew of Crete, are included for the sixth or Palm Sunday. Combined kanons by books in order to achieve as closely as possible what the Sabaites like Kosmas and John Joseph, Kosmas, Andrew and Theodore are given for weekdays of the Great week and have done (Skaballanovich 1910-1915, 1, pp. 395-396; Baumstark 1927, pp. 22-23). This Easter. A combined full kanon by Marcus the Monk with odes first to fifth and by is proved especially by the combined kanons for the Great week by the poet-composers Kosmas and Andrew with odes sixth to ninth is included for Great Saturday. A work of the two schools: the Studites have inserted their odes among the odes in the kanons of Tarasios the Patriarch is prescribed for the same day. Kanons by John the Monk, by Kosmas and Andrew. The manuscript evidences that the works of the Sabaites have Andrew and Michael are included for Easter. Combined triodia kanons by Joseph been preserved for the greatest feasts, such as Palm Sunday (by Andrew, Kosmas and and Theodore the Studite are written down for the weekdays of the Holy Bright week. John), Great week (by Kosmas and Andrew), Easter and the New Sunday (by John Kanons by John the Monk, George, Andrew, Joseph, Theodore and Theophanes are and Andrew), etc. Days between these feasts are complemented with works by the included for the New Sunday (Sunday of Saint Thomas) and next Sundays after Easter. Studites, such as photagogika. The latter are performed along with the eleven Gospel The manuscript ends with a double kanon by Theodore for Saturday before Pentecost. stichera (the eothina). It is well known that the authorship of the former is attributed to emperor Constantine vii Porphyrogenitus, and the authorship of the latter to his father, Vat. gr. 771 is among the most important manuscripts that are representative of the Emperor Leo vi the Wise. The Studites have written their works for many of the Sundays early written tradition for the repertory of movable feasts throughout the year that are and weekdays for Lent, Easter and post-Easter time. Thirdly, the Vatican manuscript collected in the book of the triodion-pentekostarion. The manuscript is important, firstly, is important as an example of how a given repertory in a local practice (in our case in terms of the preparatory season of Lent. It is comparable with the earliest triodion this one of the Grottaferrata monastery) was accepted and established. According to in Greek preserved in two Sinai manuscripts from the tenth century that are identified Bertonière (1997, p. 88) the kanon for the fourth Sunday of Lent, Λόγε Θεοῦ, was used as belonging to the same book and in all probability were written in Jerusalem: ET- at the Grottaferrata monastery, because it is revealed in four manuscripts, the origin MSsc (henceforth = MSsc) gr. 734 and 735 (Bertonière 1997 pp. 24, 30, 159; Husmann of which is linked with this monastery (one of them is Vat. gr. 771). At the time when 1975, p. 55; Follieri and Strunk 1975, p. 33). In both manuscripts, the Vaticanus and the manuscript was written various cycles of kanons existed, and their use and place the Sinaiticus, three weeks from the preparatory season starting with the Sunday of in the services has not been strictly determined. In the Vatican manuscript we find the Prodigal Son are given. These manuscripts are among the earliest ones containing a earliest four cycles of triodia kanons for weekdays of Lent: one by Theodore, one by repertory for this Sunday.5 In the Sinai manuscript this Sunday is designated as «Sunday Clement and two by Joseph with or without an acrostic of his name in the troparia of of Prodigal Son». The designation «Sunday of Prodigal Son before Meetfare Sunday», the ninth ode of the kanons. Fourthly, the manuscript is important because of the theta- what we read in Vat. gr. 771, is revealed in another early Vatican manuscript, Vat. gr. notation. The notated stichera are included in the initial part of the manuscript and 1067, a Gospel-Lectionary from the ninth century. The first Sunday of the preparatory follow in full consequence of both weekdays and Sundays for four weeks, from third season that of the Publican and Pharisee, which, as was established, was formed at Sunday vespers of Lent to Good Friday vespers (figure 7, p. 542-543). latest, is missing in these manuscripts.6 The earliest dated triodion, MSsc gr. 736 from 1027-1028, starts with the Sunday of Publican and Pharisee: the initial rubric of this After the notated stichera idiomela the feast days of Lent are repeated with kathismata, manuscript says that it is the beginning of the triodion and attributes the entire book stichera prosomoia and kanons; the stichera idiomela are not included. The first to Joseph and Theodore the Studites.7 Secondly, the Vatican manuscript is important stichera idiomela complementing the stichera prosomoia and kanons are for Lazarus because it gives evidence about the synthesis between the Sabaites and Studites in Saturday (fol. 132), then for Palm Sunday (fol. 139) and so on according to the feast compilation of the repertory of the triodion as a separate book. Michael Skaballanovich days of the calendar. The study of these stichera idiomela shows that they are different and Anton Baumstark indicate that both Theodore the Studite and his associates from from the notated ones for the same feast days in the first part of the manuscript. the Studios monastery and the monks from Italo-Greek monasteries imitate genres Besides, here they are of various kinds. For vespers, for instance, they are indicated as «stichera on Κύριε ἐκέκραξα»; stichera in stichoi complement them; stichera in stichoi complement also the morning or the Orthros stichera. It remains a question of why 5. In the Sinai manuscript this Sunday is designated as «Sunday of Prodigal Son». The designation «Sunday of Prodigal Son before Meetfare Sunday», what we read in Vat. gr. 771, is revealed in another early Vatican manuscript, Vat. gr. 1067, a notated stichera idiomela are placed at the beginning of the manuscript and separately Gospel-Lectionary from the ninth century. from kathismata, stichera prosomoia and kanons. In terms of this arrangement the 6. According to Momina (1986, pp. 112-113) the older triodion is, the less preparatory Sundays it contains. manuscript is unique: it does not have any other analogue. It could be speculated 7. The inscription with the year of writing of this manuscript is on fol. 307v. The name of the writer is mentioned «the sinful Joseph». There are notated stichera idiomela in the archaic Coislin notation on fols. 131 and 243v. I identified these whether the notated stichera idiomela were included for the first two weeks of the stichera for August 6, the Transfiguration. They could have been written later but it remains a question. Husmann (1975, p. preparatory season of Lent on the initial pages of the manuscript that are missing. 57) considers this manuscript but does not mention the notated chants in it.

540 541 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

Figure 7 Vat. gr. 771, fols. 7v-8 The end of the first part of the manuscript with stichera idiomela in theta-notation

542 543 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

As I have said, in the eleventh century, when the manuscript was written, the repertory has not been strictly established yet: any copyist could put as much items as he wanted according to his taste, education or practice preferred. At the same time the stichera idiomela were included in the notated full sticheraria consisting of menaion, triodion- pentekostarion and the oktoechos, as the comparation with manuscript A-Vn Theol. gr. 136, a sticherarion from the beginning of the twelfth century, confirms (Wolfram 1987). It is very likely that the stichera in Vat. gr. 771 were notated at the Grottaferrata monastery where the manuscript was written. The manuscript is one of the earliest witnesses of the liturgical practice in this monastery (Bertonière 1997, p. 158). Thus, it could be concluded that the theta-notation was in use at Grottaferrata in the eleventh century.

The prototype of the manuscript, however, could have been written somewhere in Constantinople or in its region. It might be linked to the school of the Studios monastery: it gives an idea about the triodion that was compiled by Theodore the Studite (Momina and Trunte 2004, p. 278). The authors included prove this, as many of them have worked at the Studios monastery. The stichera and photagogika attributed to the two emperors involved in hymnography, Leo vi the Wise and his son Constantine vii Porphyrogenitus, are another proof: it is established that the imperial works were written down and notated in manuscripts linked to the capital or its region. It is well known that the Studios monastery had an imperial status. Further, some of the hymnographers whose works are included in the manuscript may be associated with Constantinople. Michael, for instance, the author of the Easter kanon, could be identified as Michael Synkellos (760- 846), a homilist, grammarian, close to the prominent hymnographers Theodore Graptos and Theophanes Graptos, abbot of the Hora monastery in Constantinople, where he died and was sanctified (Kazhdan 1991, 2, pp. 1369-1370).8 Sergios, the author of stichera, could be identified as Sergius Confessor, born in Constantinople and died after 829; but the information about him is insufficient (ibidem, 3, p. 1880). Tarasios Patriarch, whose name is written above the work for Great Saturday, could be the Tarasios who was the patriarch of Constantinople from 784 to 806 (ibidem, p. 2011); close to him, finally, could be George, the author of kanons for Bright week: George, who died after 810, was synkellos of Tarasios (ibidem 2, p. 839). Theta signs are placed on last syllables of some of the final words of the phrases both polysyllabic likeδιό , γυνμνός, πίστις, θαῦμα, ἡμᾶς, Θεός, and monosyllabic like μέν, δέ, etc. All of these words are important in semantic meaning requiring explanation or interpretation, which is stressed by the notation. Most of them are differentiated by dots on their both sides (figure 8).

The word διό (meaning because, that is why, because of that) is among the most often notated words. The same word with a neume sign above was pointed out by Jørgen Raasted in one of the earliest known hymnographic manuscripts, US-PRu (= PRu)

8. The same Easter kanon in first mode, which in Vat. gr. 771 is ascribed to Michael, is revealed in manuscript IL-Jp Saba 83, a heirmologion from the eleventh-twelfth century in archaic Coislin notation that was converted about the fourteenth Figure 8 century into Middle Byzantine notation. In the latter manuscript the kanon is included in Saint apostle Luka’s akolouthia Vat. gr. 771, fol. 6 (Raasted 1968-1970, Pars Suppletoria, p. 54). Chants for Palm Sunday

544 545 CANTUS PLANUS 2014 KUJUMDZIEVA — THE TROPOLOGION VATICANUS GRAECUS 771

Garrett 24. The manuscript is a fragment of palimpsest of four bifolia from the second And finally, looking at the Vatican manuscript as a whole, we could say that it reveals a half of the eighth or the very beginning of the ninth century: the year 800 is accepted liturgical and hymnographical practice during the time before the establishment of the as terminus post quem when the manuscript was written (Raasted 1992, pp. 219-232; neo-Sabaitic synthesis. Without doubt, this practice was adopted in the newly baptized Jeffery 2003, p. 14). Having in mind that the theta sign is used in manuscripts stemming Slavic countries in the east. Due to the freedom of this practice, probably motivated by from different geographical places and that it is put on similar places in them in the the presence of various redactions of the Studite typikon (such as of Alexios the Studite same genre of stichera (the idiomela) speaks that it could have been a center, in which or of Evergetis, etc.), these countries were able to make their choice, selection and the theta-notation was worked out, developed and systematized. It is likely that this adoption of it to their own needs and requirements. I would stress in conclusion that notation spread from this center (Raasted 1962, pp. 302-310; Kozhuharov 1980, pp. creative rethinking, and not mechanical implementation, outlines the contribution 228-247). Such a center could have been the Saint Sabas monastery near Jerusalem of Slavic Orthodox countries to further development of Orthodox music. It was this where the great hymnographers around saint John of Damascus worked. This is an contribution that flows in the formation of a tradition regarding the acquisition of argument in favor of Oliver Strunk’s assumption (1977, p. 108) that the palaeobyzantine books, repertory, authors and overall musical culture of the Eastern Orthodox Church. Coislin notation, which contains the theta sign in its notational fund, originates from Jerusalem.

The study of Vat.gr. 771 allows us to summarize the following about it. The manuscript shows a synthesis of the two greatest Eastern hymnographic schools between the Bibliography seventh and the twelfth century, that of Saint Sabas monastery and that of Studios monastery. The synthesis is revealed in the combined repertory of authors of both Baumstark Anton (1927), ‘Denkmäler der Entstehungsgeschichte des byzantinischen Ritus’, Oriens Christianus 3/2, pp. 1-32. schools and displays a continuation of the work of Sabaites by the Studites. There is a continuous enrichment of previously created materials. The strongest direct links Bertonière Gabriel (1997), The Sundays of Lent in the Triodion: The Sundays Without a Commemoration, between Italy and Byzantium are established between the sixth-eighth and the late Roma: Pontificio Istituto Orientale (Orientalia Christiana Analecta 253). ninth-mid eleventh century when South Italy was a province of Byzantium (Nardini Devresse Robert (1950), Codices 604–866, Roma: Biblioteca Apostolica Vaticana (Codices Vaticani Graeci). 2007, p. 163). Works of Sabaites, as I have said, are preserved for the greatest feasts; Ermatinger Charles J. ed. (1993), Guide to Microfilms of Vatican Library Manuscript. Codices Available works of Studites are for rest of the feasts. There are some works of the Studites for for Study in the Vatican Film Library at Saint Louis University, Saint Louis: The Pius xii Memorial great feasts but they are combined with those of Sabaites. The order of the services is Library. strictly preserved. Each service, however, could contain pieces by different authors, Follieri Enrica - Strunk Oliver eds., Triodium Athoum. Cod. Monasteri Vatopedi 1488, Pars Principalis et which speaks of some freedom in the choice of repertory for a given feast. There is Pars Suppletoria, Copenhagen: Munksgaard (Mmb, Série Principale 9). freedom, also, in terms of what should be notated and how. In all probability it was Frøyshov Stig S. (2013), Greek Hymnody in John R. Watson - Emma Hornby eds., The Canterbury Dictionary a writer’s choice: the writer could decide what to include in a given manuscript. of Hymnology, Norwich: Canterbury Press http://www.hymnology.co.uk/g/greek-hymnody (9 Some principles concerning especially the theta-notation, however, are revealed: for November 2015). instance, words in chants for great feasts that are important in meaning are notated Husmann Heinrich (1975), Ein syro-melkitisches Tropologion mit altbyzantinischer Notation, Sinai Syr. 261, most. Obviously, this method of notation became the norm and it was this method that 2 vols., Wiesbaden: Harrassowitz (Göttinger Orientforschungen 9). was perceived not only in Byzantium but also in the newly baptized Slavic countries Jeffery Peter (1991), ‘The Sunday Office of Seventh Century Jerusalem in the Georgian Chant Books. A in a very early time. The Vatican manuscript raises the question also of when and Preliminary Report’, Studia Liturgica 21, pp. 52-75. where the repertory of the Eastern Church started to be notated: in Jerusalem and respectively in the school of Sabaites in the eighth century, or later on, by the end of – (2001), The Earliest Oktoechoi: The Role of Jerusalem and Palestine in the Beginning of Modal Ordering in Peter Jeffery ed., The Study of Medieval Chant: Paths and Bridges. East and West. In Honor of the tenth or the first quarter of the eleventh century when the work of Studites had Kenneth Levy, Cambridge: Boydell, pp. 147-211. already flourished. Judging by the earliest manuscripts like PRu Garrett 24 and the – (2003), A Window of the Formation of the Medieval Chant Repertories: The Greek Palimpsest Fragments menaia and triodia in which separate chants or parts thereof are notated, such as in Princeton University MS Garrett 24 in The Past in the Present. Papers Read at the Fourth Meeting MSsc gr. 569, 581, 613, 736, etc., the last assumption seems more likely: the notated of the Ims Intercongressional Symposium and the tenth Meeting of the Cantus Planus (Budapest and books (of the type of the sticherarion and heirmologion) were compiled later and they Visegrád 2000), 2 vols., Budapest: Liszt Ferenc Academy of Music, 1 pp. 1-21. could be linked with the formative liturgical and hymnographical work of Studites. Kazhdan Alexander ed. (1991), Oxford Dictionary of Byzantium, 3 vols., Oxford: Oxford University Press.

546 547 CANTUS PLANUS 2014

Silvia Tessari Kujumdzieva Svetlana (2008), Belezhki i komentari za njakoi ranni himnografski rakopisi ot Vaticana [Remarks and Commentaries About Some Early Hymnographic Manuscripts from Vatican] in Elena Toncheva - Svetlana Kujumdzieva - Asen Atanasov eds., Bogosluzhebmite knigi - poznati i nepoznati The Byzantine musical manuscripts of the Veneto region: [Liturgical books known and unknown]. Sofia: Omofor, pp. 43-155. overview of a current research project. – (2011), The Tropologion Vaticanus Graecus 2008 in Nina-Maria Wanek ed., Psaltike. Neue Studien zur Byzantinischen Musik: Festschrift für Gerda Wolfram, Wien: Praesens, pp. 187-205. One example (Bassan. gr. 34B19)

– (2013), Rannite osmoglasnici. Izvori, bogosluzhenie i pevcheski repertoar - po rakopisi do xiii vek [The early oktoechoi. Sources, liturgy and chant repertory - Based on the sources until the thirteenth century], Sofia: Cyrillo-Methodian Research Centre, Bulgarian Academy of the Sciences. In the libraries of the Veneto region, Byzantine neumated manuscripts, dated from the ninth century until the eighteenth century, though quite numerous ad significantly Kozhuharov Stefan (1980), ‘Palaeographski problemi na tita notacijata v srednovekovnite rakopisi ot xii- xiii vek’ [Palaeographical Problems of theta notation in medieval manuscripts from twelfth-thirteenth relevant for the development of music history and Byzantine musical palaeography – centuries], Slavjanska paleographia i diplomatika 1, pp. 228-247. especially the Nanian collection now at the Marciana Library (Romanou 2006; 2009,

Metreveli Elena P. - Čankieva Caca A., Hevsuriani Lili eds. (1980), Udzvelesi Iadgari, Tbilisi: Mecniereba, pp. 115-117, 119; Tessari 2013) – have never been fully studied and described. Certain 1980. collections, however, like that kept at the Istituto Ellenico di Studi Bizantini e post- Bizantini of Venice, or manuscripts of specific origin (as those from 1566-1669 Crete) Momina Maia A. (1986), ‘O proizhozhdenii grecheskoj triodi’ [About the origin of the Greek triodion], Palestinskij sbornik 28, pp. 112-113. can count now on recent publications (Stathes 2007; Giannopoulos 2004). Aim of this paper is a brief presentation of some results derived from a biennal post-doc research Momina Maia A. - Trunte Nikolaos H. (2004), Triodion und Pentekostarion. Nach slawischen Handschriften des 11.-14. Jahrhunderts, Paderborn-München-Wien-Zürich: Nordrhein-Westfälische Akademie der project I am conducting at Padua University (Department of Cultural Heritage, Wissenschaften. supervisors: Anna Pontani, Antonio Lovato), devoted to the first systematic cataloguing of these Byzantine and post-Byzantine sources, both those penned in the so-called Nardini Luisa (2007), ‘Aliens in Disguise: Byzantine and Gallican Chants in the Latin Liturgy’, Plainsong and Medieval Music 16, pp. 145-172. ecphonetic notation, i.e. Greek lectionaries, and those with melodic notation, i.e. hymn books of different types.1 My research focused on the following libraries:2 Parenti Stefano (2000), Mesedi – Μεσώδιον in Hans-Jürgen Feulner - Elena Velkovska - Robert Taft eds., Crossroad of Cultures. Studies in Liturgy and Patristics in Honor of Gabriele Winkler, Roma: Pontificio Istituto Orientale (Orientalia Christiana Analecta 260), pp. 543-556. Verona, Biblioteca Capitolare (I-VEcap): one manuscript with melodic notation Raasted Jørgen (1962)‚ ‘A Primitive Palaeobyzantine Musical Notation’, Classica et Medieavalia 23, pp. 301‑310. Padua, Biblioteca Antoniana (I-Pca): – (1992), ‘The Princeton Heirmologion Palimpsest’, Cahiers de l’Institut du Moyen-Age Grec et Latin 62, two fragments with ecphonetic notation pp. 219-232. Padua, Biblioteca Universitaria (I-Pu): – ed. (1968-1970), Hirmologium Sabbaiticum Codex Monasterii S. Sabbae 83, Copenhagen: Munksgaard (Mmb, Série Principale 8). five manuscripts of Cretan origin, with melodic notation

Renoux Charles (2001), Jerusalem dans le Caucase: Anton Baumstark verified in Robert Taft - Gabriele Bassano del Grappa (Vicenza), Biblioteca Civica (I-BDG): Winkler eds., Comparative Liturgy. Fifty Years After Anton Baumstark († 1948). Acts of the one fragment with melodic notation international Congress (Rome, 25-29 September 1998), Roma: Pontificio Istituto Orientale (Orientalia Christiana Analecta 265), pp. 305-321. Venezia, Biblioteca Nazionale Marciana (I-Vnm): Skaballanovich Mikhail (1910-1915), Tolkovij Typikon [Interpreted Typikon], 3 vols., Kiev: GT Korczak- nineteen manuscripts/fragments with ecphonetic notation, twelve with melodic Nowicki. notation and other two with only theoretical sections.

Strunk Oliver (1977), Essays on Music in the Byzantine World, New York: Norton.

Touliatos-Banker Diane (1987), ‘Checklist of Byzantine Musical Manuscripts in the Vatican Library’, 1. I would like to thank – for the help and the guidance – particularly Sandra Martani (University of Pavia/Cremona, Manuscripta 31/1, pp. 22-28. Conservatory of Mantova), whose researches on Byzantine music in Italy have been the first indispensable start for the project. Wolfram Gerda ed. (1987), Sticherarium Antiquum Vindobonense, Wien: Verlag der österreichischen 2. Selected bibliography about musical manuscripts of the Veneto region: Verona, Formentin 1980-1981, pp. 185-186; Akademie der Wissenschaften (Mmb, Série Principale 10). Piazzi 1994, pp. 98-99; Marchi 1996, p. 219; Arnesano 2005, p. 139; 2010, pp. 87 and 92. Padua and Venice, Giannopoulos 2004, pp. 671-705 and pp. 741-750. Venice: Papageorgios 1907; Tessari et al. 2014.

548 549 CANTUS PLANUS 2014 TESSARI — THE BYZANTINE MUSICAL MANUSCRIPTS OF THE VENETO REGION

In this paper, I will briefly present one example of my work, taking into account thirteenth-century, provincial, eastern Mediterranean product. The dating is also – of the manuscripts with melodic notation – the older source I could identify, a supported by a subscription, hitherto escaped to the attention of the few scholars parchment fragment of sticherarion used as posterior counterguard of a Greek Gospel who took into account the manuscript. Paolo M. Tua (1931) and Elpidio Mioni (1965) lectionary, which has yet to be described appropriately, kept in the Civic Library of considered it a fifteenth-century lectionary, while Kurt Aland (1994) placed it in the Bassano del Grappa, a town situated northeast of Vicenza: I-BDG Manuscript 1087, fourteenth century. The subscription, at fol. 80v, is realized with light brown ink, coll. 34 B 19 Tua (henceforth B), thirteenth century, parchment codex, fols. 136, 250 almost totally faded: «μηνὶ αὐγούστου εἰς τὰς κε΄ ἡμέρα [ut videtur] κυριακὴ ἰνδικτιῶνος × 190 (Tua 1931, p. 205; Mioni 1965, p. 17; Aland 1994, p. 336). <…> ἕτους ςψξ΄», that is: «25 August, on Sunday, indiction [no numbers visible], year 6760» [annus mundi = 1252 d. C.]. The manuscript as a whole The manuscript is a complete Greek Gospel Lectionary (ℓ1910 ℓesk) not provided Distinctive script. The beginning of each section of the Lectionary is marked with with ecphonetic notation. It bears an original binding made of thick wooden a rubric in slender epigraphic majuscule script written in carmine ink. Marginal boards covered with dark-brown leather. The decorations of the two covers are annotations: the word ἰαματικόν, often incorrectly written, is penned on the upper diverse: on the front, concentric frames give way to a central rhombus with cross margin by a later hand where the reading from the Gospel describes Jesus’ miracles. in the centre (roughly made) and four little roses on the corners. On the back the Many hymnographic marginalia are present: at fol. 8v there is one kathisma for the geometrical ornamentation is more accurate, with little roses in the middle and Virgin in the fourth plagal mode, incipit: «Παναγία Παρθένε Μήτηρ Θεοῦ, τῆς ψυχῆς on the corners. On the spine of the book there is the modern title: Evangelarium μου τὰ πάθη τὰ χαλεπά». At fol. 111v we can read: «ὁμοουσίου Τρια[…] πυμὴν τῶν […] Graece in membranis. The manuscript is made of a well-preserved thick grey-yellow προβάτων», probably a hymnographic text, hitherto not identified. At fol. 136v, at the parchment, of modest quality. The difference between the flesh side and the hair end of the Lectionary, there are liturgical rubrics almost entirely faded (prokeimena side is easily discernible (what is known as Gregory’s law is, therefore, respected). and stichera indications, along with their musical mode). They are written by a later There are clear speckled traces of hair follicles and some circular holes. The original hand in black ink. Greek numbering of the seventeen quires (all quaternions), realized with black ink, can be seen on the upper right corner of the first leaf of each quire, but it is often Ornamentation. There are little crosses at the top of some leaves (fols. 23v, 65v, 94v, cut. There is no modern foliation. The manuscript was written by a single scribe, 101v, 105, 109, 125). The ornamentation (monochrome in general) includes bands who used a deep black ink and penned from twenty-six to twenty-nine lines per and - headpieces. See the red and black band with hearts and vegetal motifs at fol. leaf, on two columns. The minuscule script is peculiar, with a rapid ductus, and π 1. At the beginning of the readings from Matthew (fol. 26), on the other hand, there a round stylization, with cursive elements. Some letters are larger than the ones is a weave-like polychrome band, in which a pale yellow/beige ink is used. Initials, near them (e.g. majuscule Ε, Θ, Κ). Note triangular α, heart-shaped β, round and sometimes very elongated (five-twelve lines of writing) and traced mainly with a flat θ, similar to the Biblical majuscule Θ. The vertical trait of the minuscule κ is single line, penned in carmine ink, show also the motives of the blessing hand and curved, nearly semi-circular. Note also τ higher than the other letters, sometimes of the hand which holds a cross. The ornamental motifs are few: the horizontal trait μεν distendu (Canart 1981, pp. 66-67), majuscule Ε, big and lunate, which joins the of has spiral rings at the two ends. A zoomorphic initial (fish-shaped omicron) is at following letter by way of low pseudo-ligatures. Breathings and accents also show τ 109v. At fol. 71, on a large part of the right column, erased because the text had been peculiar features: the circumflex accent is often written on the last consonant of the erroneously copied twice, there is a drawing (probably subsequent to the writing of word. The grave accent is almost horizontal. The use of breathings and accents is the manuscript) of two fantastic animals (panthers?) with mottled hair, holding a very incorrect, with manifold erroneous segmentations of the words. The graphic thin column surmounted by a plume in one paw. Other traces of writing are visible features of the codex are very similar to those of the main scribe of the manuscript behind the drawing. GB-Lbl Add. MS 19460 – though penned in a less cursive handwriting – (Gregory- Aland ℓ192; Richard 1952, p. 35; Aland 1994, p. 230; Catalogue of Additions, p. 243) dated to the thirteenth century on palaeographical base and collocated by P. History of the manuscript. There is a note of possession at fol. 134v: «Τὸ παρὸν Canart (1981, pp. 65-66) and G. Prato (1994, p. 61) among a group of manuscripts ἅγιον ˙ ἐβακγέλιον ˙ ὑπἅρ(χει) ἐμοῦ ˙ Ἰωακὴμ ἰερομονάχου ˙ τὸ πίκλην λίβας», that is of likely Greek-oriental origin: GB-Ob Arch. Selden. B 54, fols. 155-217v, and B 56; «this saint Gospel belongs to me, Ioakeim hieromonk nicknamed Libas». A «Ἰωάννης GB-Lbl Royal. 1.D.II; V-CVbav Vat.gr. 859, 1616, 1840, 1886, 2319 (Cataldi Palau Δημήτριος» appears on fol. 135. The name is penned in the same brown ink used for and D’Agostino 2010, pp. 158-160). The Bassanensis is therefore, is my opinion, a the realization of the dated subscription.

550 551 CANTUS PLANUS 2014 TESSARI — THE BYZANTINE MUSICAL MANUSCRIPTS OF THE VENETO REGION

Front counterguard. Only a parchment fragment of irregular shape survives (almost There are twenty double lines per leaf (the neumated line is as usual written above triangular, six cm wide and two cm high). On the first line, we can read «ενον οὐδὲν», the text). The script is a mixed minuscule of the traditional type (majuscules, rather on the second «τὸ δὲ κατορθούμε», and therefore we can assume that it is a fragment infrequent: Ζ, Η, Θ, Κ, Λ, Φ). The letters, written in dark brown ink, are not provided from the homily on the Epistle to the Hebrews of Saint John the Chrisostome. This is with accents and breathings (as quite normal in musical manuscripts); of small and the complete sentence: «Καίτοι τὸ γενόμενον οὐδὲν ἦν, τὸ δὲ κατορθούμενον μέγα. τὸ μὲν square module, they have short ascendings and descendings and vertical axes. The γὰρ γενόμενον αἷμα, τὸ δὲ κατορθούμενον σωτηρία καὶ ἔμφραξις καὶ κώλυμα ἀπωλείας» cola are divided by high dots, while the end of the stichera is marked with :- . Cursive (Pg lxiii, col. 185, line 32 ff.). The original leaf was probably written on two columns, ligatures are generally avoided, as usual when each syllable has to be accompanied each of them seven cm wide. The script is a Perlschrift like writing (eleventh century?). by its respective neume or neumatic combination (but see ἄγγελοι, tenth line, on the recto, with elongated epsilon made of a single roughly semi-circular trait tied with the The counterguard with musical notation basis of the following lambda). Abbreviations, too, are absent, with the exception of Byzantine melodic notation can be found only on the posterior counterguard. It is still the tachygraphic καί. In the leaf there are no rubrics, nor authorial ascriptions. The not damaged and partially liftable, though attached to the board by one of the two initials, of modest size and adorned with simple hooks, are written in red ink. The leather strips, and derives from a Greek Sticherarion (musical manuscript containing initial martyriae were written in the same ink, but on the recto they are completely lost, the – mainly idiomela – stichera: brief chants commemorating the saints and the other because of the fading of the red pigment. The notation belongs to the so-called Coislin immovable and movable feasts of the liturgical calendar, to be sung throughout the type, in its fully developed stage (Coislin v), whose degree of diasthemacy is quite year during the office hours in alternation with or immediately after psalm or other high, while not reaching the perfect diasthemacy of the middle Byzantine notation Scriptural verses) penned in the so-called Paleobyzantine melodic notation. This (Strunk 1966, Pars suppletoria p. 33; Floros 1970,1, pp. 311-326). In particular, both notation, not yet diasthematic, spread in the Byzantine Empire, in two main types and the shape of the neumes (penned in the same ink as the text, and the same stylus) and in different developmental phases, between the tenth to the last quarter of the twelfth the melodic line can be almost completely overlapped with those of the Sticherarium century, when the so-called middle Byzantine notation, diasthematic, appeared. The of the Austrian National Library in Wien (A-Wn Theol. gr. 136, henceforth Vind.136), counterguard, made of parchment of good quality, maintains the original dimensions dated to the twelfth century (Wolfram 1987). Main features: (230 × 170). The Palaeobyzantine neumes and the Greek text can be easily read both all syllables bear notation (while in the more archaic phases many syllables were let on the recto and on the verso. There are six stichera in honour of the Dormition of the without neumes); Virgin (15 August), all of them widely known in Greek Orthodox tradition and in the Churches that follow the Byzantine rite, and usually copied at least in all the sticheraria analytical manner of writing the thematismoi, that is: the melismas are marked not manuscripts from around 1050 onwards; they belong to the so-called Sav: Abridged only by the letter theta of the Greek alphabet, which had been used with this purpose Version of the Sticherarion (Strunk 1977, p. 107; Troelsgård 2003). Here the incipit: since the first dawn of Byzantine notation, but also with a series of neumes serving to define the melody in a more precise way; Ὅτε ἐξεδήμησας Θεοτόκε, fourth authentic mode (Sav no. 722) diasthematic boundaries of certain conjunctures and ligatures (e.g. the mega beginning from: τῆς γῆς Ποιητήν, δοξολογίαις kratema plus kentema or elaphron, bareia plus apostrophos, ison with oxeia under Δεῦτε φιλεόρτων τὸ σύστημα, δεῦτε καὶ χορείαν στησώμεθα, first plagal mode or above it, bareia with kentema); (Sav no. 724) chamele and hypsele are present, but their shape does not have a precise diasthematic Ἄισατε λαοί, τῇ Μητρὶ τοῦ Θεοῦ ἡμῶν ᾄσατε, first plagal mode S( av no. 723) value (fifth downwards and upwards);

Τῇ ἀθανάτῳ σου κοιμήσει, Θεοτόκε Μήτηρ τῆς ζωῆς, second plagal mode neumes signifying, in middle Byzantine notation, a third or a fourth downwards do (Sav no. 725) not yet have this specific value: e.g. three neumes indicating, in middle Byzantine notation, a fourth downwards are here (eighteenth line, on the recto, see example 3) , second plagal mode Δεῦτε τὴν παγκόσμιον κοίμησιν τῆς παναμώμου Θεοτόκου written one after the other (a clearly impossible melodic line), while in later sources (Sav no. 726) the melody descends stepwise;

Ὅτε ἡ μετάστασις τοῦ ἀχράντου σου, second plagal mode (Sav no. 727), the initial martyriae (indicating the musical mode of each composition with a letter: last words: ἱκέτευε ἐκτενῶς. α, β, γ, δ for the four authentic modes and preceded by «πλ». for the plagal modes,

552 553 CANTUS PLANUS 2014 TESSARI — THE BYZANTINE MUSICAL MANUSCRIPTS OF THE VENETO REGION

often in conjunction with neumes showing the starting note of each melody) are Please note that the opening melodic line (transcribed as CDD from the middle neumated. They are visible only on the versus and are written in red carmine ink. Byzantine sources) is repeated identically at the beginning of the second colon in all E.g. the neumes of the martyria of second plagal mode which opens the sticheron the considered sources (always on the word δεῦτε), but the notation is different: in Δεῦτε τὴν παγκόσμιον κοίμησιν are: dyo apostrophoi plus klasma. the first case dyo apostrophoi, oxeia with kentema, and diple, while in the second one the kentema is omitted, like in Vind136. Generally speaking, the comparison between The abovementioned features allow us to date the counterguard at the twelfth century, B and Vind136 shows the oscillation between kentema and kentemata, and between overcoming the previous hypothesis formulated by Mioni (fifteenth century, coeval oxeia and petaste. B, e.g., has a petaste on the stressed syllable of ἐνατενίζοντες, where with the manuscript). In the following lines, we will analyse some melodic features of all the other manuscripts bear an oxeia. B has a beginning parakletike on the first each sticheron in B comparing them with these sources: syllable of Σήμερον, where Vind136 has a simple oligon plus kentema above (forth upwards) and the other manuscripts (in middle Byzantine notation) a fifth upwards A-Wn Theol. gr. 136, twelfth century (Wolfram 1987); henceforth Vind136 (see the first neume shown in example 2). The initial martyria is missing, as always on the recto. There are medial signatures of first and first plagal mode, without neumes, A-Wn Theol. gr. 181, a. 1221 or 1217 (Høeg, Tllyard and Wellesz 1935); marked by a horizontal stroke on the letter α and written in the same ink of the text henceforth Vind181 and of the neumes (brown). Megala semadia and melismatic passages: see example I-Ma A 139 sup., a. 1342 (Perria and Raasted 1992); henceforth A 2 for the thematismos on the text σήμερον] γάρ. The same neumatic figure is written above the text ζω]ῆς. The thematismos, in both cases preceded by the signature of the Μηναῖα τοῦ ὅλου ἐνιαυτοῦ, (1901, 6); henceforth MR first authentic mode and immediatelyfollowed by the medial signature of first plagal mode, corresponds to an ouranisma formula in A with a non-formulaic ending. Other The transcriptions refer exclusively to the neumes of A. megala semadia are: xeron klasma and katabasma (like Vind136), kratema, kylisma, seisma ii (apostrophos+piasma+petaste). 1. Sticheron Δεῦτε φιλεόρτων τὸ σύστημα, δεῦτε καὶ χορείαν στησώμεθα, first plagal mode [Theophanes Protothronos, ninth century] Example 2. From the sticheron Δεῦτε, φιλεόρτων τὸ σύστημα, fifth colon, first plagal mode Text: την τεξαμενην B; Vind181; A την δεξαμενην Vind136 B τὴν τετοκυῖαν MR

Position. After Ὅτε ἐξεδήμησας, Θεοτόκε. Before ᾌσατε λαοί, τῇ Μητρὶ τοῦ Θεοῦ ἡμῶν ᾄσατε; the same position in Vind136, Vind181, while in A it is copied after ᾌσατε λαοί, τῇ Μητρὶ τοῦ Θεοῦ ἡμῶν ᾄσατε. Differences between B and Vind136: almost negligible Vind136 differences. See the first colon in example 1. Vind181 Example 1. Incipit of the sticheron Δεῦτε, φιλεόρτων τὸ σύστημα, first plagal mode A

B G G b a G G a ba ba G a E F a bc a Ga Ga E F Ga a

Δε τε φιλεόρτων τὸ σύστημα

Vind136 2. Sticheron Ἄισατε λαοί, τῇ Μητρὶ τοῦ Θεοῦ ἡμῶν ᾄσατε, first plagal mode, Sav no. 723 [Theophanes] πλα´ The text is identical to the one printed in the liturgical books. The position of the sticheron in the section dedicated to the Dormition of the Virgin is after Δεῦτε, φιλεόρτων τὸ σύστημα. The neumatic line shows only minor differences with that of Vind136.

554 555 CANTUS PLANUS 2014 TESSARI — THE BYZANTINE MUSICAL MANUSCRIPTS OF THE VENETO REGION

B seems to be more advanced in the road towards diastemacy (e.g. the bareia with between B and Vind136: there are only negligible differences. An apostrophos in B apostrophos below, instead of the single bareia of Vind136. On the contrary, B has the corresponds with an ison in Vind136, where the middle Byzantine notation of Vind181 simple petaste instead of petaste with dyo kentemata of Vind136). B does not employ and A bears a third downwards. Megala semadia and melismatic passages: for the the dot after kouphisma that we find in Vind136. A melodic difference can probably be Marian invocation Χαῖρε, Vind181 and A show the so-called chairetismos: a melismatic envisaged at the end of the sticheron, on the word εἰρήνην (see example 3). passage that we find exactly in the same form in that colon of Ioannes Koukouzeles’ didactical song Mega Ison dedicated to the chairetismos (see e.g. fol. 4 of the ms. GR- An 2458, dated 1336 and thus contemporary with A). The ‘Paleobyzantine’ mss. B Example 3. Last colon of the sticheron Ἄισατε λαοί, τῇ Μητρὶ τοῦ Θεοῦ ἡμῶν ᾄσατε, first plagal mode and Vind136 express the melisma with nearly the same notation (Vind136 has the strangismata in addition): see example 4. B and Vind136’s chairetismos is identical to the notation of this melisma in the athonite ms. GR-AOml Γ 74 (tenth century, penned B [ … ] in the so-called Chartres notation). It appears to be a sort of simplification of thema haploun neumatic conjunctions (as can be found e.g. in ms. GR-AOml Γ 67, tenth century). It is quite normal – as demonstrated by the research conducted by Annette Jung (1999, pp. 26-29) – to find confusion between the megala semadia of chairetismos and thema haploun in Palaeobyzantine sources, as also shown by B. This confusion Vind136 was later eliminated, when, after the transition to the round notation, the distinction between the two melismas was re-established. The chairetismos figure in Vind181 and Vind181 A is closed by a thes kai apothes passage (the melody in A: FGFG), expressed in A not only with the neumatic sequence, but also with the respective mega semadion ΘΘ. On Μεθ’ὧν, which opens the last syntactic period, there is an usual form of thematismos, A with an ascending/descending (transcription of A’ neumes: DGadcb).

e f a G F C D EF D EF a G a F E F G d F D D Example 4. Eleventh colon of the sticheron Τῇ ἀθανάτῳ σου Κοιμήσει, second plagal mode

The initial martyria is missing. Medial signature of first and first plagal mode are placed at the beginning and at the end of the melismatic passage σήμερον γάρ (see B below), so that it is the only part of the sticheron to be chanted in the first authentic mode, while in Vind181 the plagal mode comes back later, one colon after, at the end of τὴν ὁλόφωτον ψυχὴν αὐτῆς. Vind136, on the contrary, has no medial signatures in this sticheron; A opens the ‘authentic’ section before σήμερον γάρ, but does not indicate other medial signatures after it. Megala semadia and melismatic passages: the only Vind136 melismatic passage is placed again on the text σήμερον γάρ, which quite frequently in Byzantine hymnody (as also shown in this sticheron and in the previous one) marks the separation between the exhortative section (with the imperative mode, Vind181 often preceded by δεῦτε) and the narration of the core of the feast: here the Dormition of the Virgin. The same thematismos shown above is used. In Vind181 and in A the melisma is an ouranisma followed by a figure with xeron klasma. A

3. Sticheron Τῇ ἀθανάτῳ σου κοιμήσει, Θεοτόκε Μήτηρ τῆς ζωῆς, second plagal mode (Sav no. 725) [John Monachos, seventh/eighth century] G F Ga b a Gf G E F G F G E F D G G a b b Text: no differences in comparison with the printed liturgical books. Differences

556 557 CANTUS PLANUS 2014 TESSARI — THE BYZANTINE MUSICAL MANUSCRIPTS OF THE VENETO REGION

4. Sticheron Δεῦτε τὴν παγκόσμιον Κοίμησιν, second plagal mode the edge of the sheet. There are no medial signatures. The differences between B and (Sav 726) [Anatolios] Vind136 are again unimportant, but cf. the oligon on the second, unstressed syllable, of Text ἀτενίζοντες in B and A instead of petaste or petaste with dyo kentemata in Vind136 and εορταζομεν B et alii | ἑορτάσωμεν MR Vind181. Furthermore, on the final syllable ofσὺν δάκρυσιν, B has a xeron klasma and a επι γης συγκαλουνται B et alii | γηγενεῖς συγκαλοῦσιν MR katabasma, while Vind136 has a descending progression marked by three apostrophoi, ασυγχητω B | ἀσιγήτῳ MR et alii like in the middle Byzantine notation of Vind181 and A. The hymn, a πεντηκοστάριον, χαιρε B et alii | χαίροις MR is rich in melismatic passages (e.g. on the words: Ὁ δὲ Πέτρος, σὺν δάκρυσιν, ἐβόα The initial martyria of the second plagal mode is written in red. Megala semadia and σοι, Ἀλλ’ὦ ἄχραντε, ἱκέτευε ἐκτενῶς), not always easily identifiable with proper names melismatic passages: at the beginning, Δεῦτε is sung with the chairetismos formula in on the base of their specific neumatic chain. Annette Jung (1999, p. 65) describes the both A and Vind181. The Palaeobyzantine sources show the same neumes conjunctions «long melismas» of this sticheron copied in A as: «non-formulaic, with choreuma, of the abovementioned Χαῖρε (simplification of thema haploun formula). Three Hail! heteron homoion var., + ouranisma + non-formulaic, synagma + Amargianakes 51Θ». constitute the main part of the hymn. Note that the second one, addressed to the choir On Ἀλλ’ὦ ἄχραντε, ἱκέτευε ἐκτενῶς, Vind181 and A show – as stated above – a melisma of disciples, is χαῖρε in the manuscripts, and χαίροις in the printed liturgical books, composed by a thematismos, a synagma formula (in both cases expressed by the being χαῖρε reserved only for the Theotokos. The formulas used in the Paleobyzantine mega semadion) and the Amargianakes formula 51Θ. The Palaeobyzantine notated sources can be seen in example 5. The middle Byzantine sources, for the firstχαῖρε , manuscript B – insofar as the poor state of preservation of those very last lines of the show a standard chairetismos followed by two different cadences. Χαίροις in Vind181 sheet permits – and Vind136 have almost identical notation (see example 6). Although is chanted with an ouranisma formula, followed by a thes kai apothes conclusion. In a synagma formula (together with the respective mega semadion), as seen above, is A, on the contrary, there is a chairetismos and the final thes kai apothes, marked by penned in the middle Byzantine sources, no synagma figures comparable to that we the mega semadion written in red. For the last Χαῖρε, a normal chairetismos formula find e.g. in the manuscript GR-AOmlΓ 67 (tenth century, Chartres notation list of signs) is used in Vind181 and A, followed by the thematismos marked 51α by Amargianakes are used in B and Vind136 (Coislin notation). (1977, pp. 74-79, 177, 181, 241).

Example 6. Eleventh colon of the sticheron Ὅτε ἡ Μετάστασις τοῦ ἀχράντου σου σκήνους, second plagal mode Example 5. The Hail! sections of the sticheron, second plagal mode

B B = Vind136

Χα ρε

Vind136 B

Χαίροις Vind181

Vind136 A

5. Sticheron Ὅτε ἡ Μετάστασις τοῦ ἀχράντου σου σκήνους, second plagal mode D G a d c b d c d c ba b db c ba cbaG G b a a GF EF G [Byzantios] The initial martyria of this sticheron is not easily readable, because it is located near

558 559 CANTUS PLANUS 2014 TESSARI — THE BYZANTINE MUSICAL MANUSCRIPTS OF THE VENETO REGION

Bibliography Prato Giancarlo (1994), Studi di paleografia greca, Spoleto: Fondazione Centro italiano di studi sull’alto medioevo. Agosto Mauro (forthcoming), Il cod. gr. (palimps.) CXX (110) della Biblioteca Capitolare di Verona: annotazioni paleografiche e musicali in Atti del Convegno Internazionale di Studi «Padre Lorenzo Romanou Katy (2006), ‘Σπάνια χειρόγραφα για την ελληνική μουσική. Απόσπασμα από το Storia della musica Tardo, la musica bizantina e gli Albanesi di Sicilia», Palermo, 7-9 Marzo 2008. του Giovanni Battista Martini’ [Manoscritti rari sulla musica greca. Estratto dalla Storia della musica di Giovanni Battista Martini], Πολυφωνία 9, pp. 126-140. Aland Kurt ed. (19942), Kurzgefasste Liste der griechischen Handschriften des Neuen Testaments, Berlin-New York: De Gruyter. – (2009), Serbian and Greek Art Music, Chicago: University of Chicago Press.

Amargianakes George (1977), ‘An Analysis of Stichera in the Deuteros Modes. The Stichera Idiomela for Stathes Gregorios Th. (2007), ‘Ελληνική ψαλτική τέχνη. Τα χειρόγραφα βυζαντινής μουσικής του Ελληνικού the Month of September in the Modes Deuteros, Plagal Deuteros, and Nenano Transcribed from the Ινστιτούτου Βενετίας’, Θησαυρίσματα 37, pp. 9-59. Manuscript Sinai 1230 (A.D. 1365)’, Cahiers de l’Institut du Moyen-Âge grec et latin 22-23, pp. 1-263. Strunk Oliver (1966), Specimina Notationum Antiquiorum. Pars Principalis et Pars Suppletoria, 2 vols., Arnesano Daniele (2005), ‘Il «Copista del Digenis Akritas». Appunti su mani anonime salentine dei secoli Copenhagen: Munksgaard (Mmb, Série Principale 7) XIII e XVI’, Bizantinistica. Rivista di studi bizantini e slavi 7, pp. 135-158. – (1977), Essay on Music in the Byzantine World, New York: Norton. – (2010), Manoscritti greci di Terra d’Otranto. Recenti scoperte e attribuzioni 2005-2008 in Daniel Galadza - Nina Glibetić - Gabriel Radle eds., Toxotes. Studies for Stefano Parenti, Grottaferrata: Tessari Silvia (2013), Cirillo Martini and the First Steps Towards the Discovery of Byzantine Music Theory Monastero Esarchico di Santa Maria di Grottaferrata, pp. 63-101. in 18th-Century Italy in Nicholas Giannoukakis ed., Conference Proceedings. Fourth International Conference of the American Society of Byzantine Music and Hymnology “Byzantine Musical Culture” Canart Paul (1981), ‘Les écritures livresque chypriotes du milieu du IXe siècle au milieu du XIIIe et le style (Peania, 10-13 October 2013), Pittsburgh: Asbmh, pp. 1-45 (forthcoming). palestino-cypriote «epsilon»’, Scrittura e civiltà 5, pp. 17-76. Tessari Silvia - Martani Sandra - Lugato Elisabetta - Marcon Susy - Braides Orsola (2014), Cantus Planus. Cataldi Palau Annaclara - D’Agostino Marco (2010), Manoscritti provenienti dalla Tessaglia a Milano in Notazione musicale bizantina in codici marciani. Catalogo della mostra, Venezia: Levi. Marco D’Agostino - Paola Degni eds., Alethes Philia. Studi in onore di Giancarlo Prato, Spoleto: Fondazione Centro italiano di studi sull’alto medioevo (Collectanea 23), pp. 93-167. Troelsgård Christian (2003), ‘A List of Sticheron Call-Numbers of the «Standard Abridged Version of the Sticherarion». Part I (The Cycle of the Twelve Months)’, Cahiers de l’Institut du Moyen-Âge grec et latin Catalogue of Additions to the Manuscripts in the British Museum, 1848-1853 (1868), London: British Museum. 74, pp. 3-20.

Floros Constantin (1970), Universale Neumenkunde, 3 vols., Kassel-Wilhelmshöhe: Bärenreiter. Tua Paolo Maria (1931), Bassano del Grappa. Biblioteca civica in Albano Sorbelli ed., Inventari dei manoscritti delle Biblioteche d’Italia 50, Firenze: Olschki, pp. 5-350. Formentin Maria Rosa (1980-19812), ‘I palinsesti greci della Biblioteca Nazionale Marciana e della Capitolare di Verona’, Δίπτυχα 2, pp. 144-186. Wolfram Gerda ed. (1987), Sticherarium Antiquum Vindobonense. Codex Theol. gr. 136 Bibliothecae Nationalis Austriacae phototypice depictus, 2 vols., Wien, Verlag der österreichischen Akademie der Giannopoulos Emmanouel (2004), Ἡ ἄνθηση τῆς ψαλτικῆς τέχνης στὴν Κρήτη (1566-1669) [La fioritura Wissenschaften (Mmb, Série Principale 10). dell’arte psaltica a Creta], Athen: Institute of Byzantine Musicology.

Høeg Carsten - Tillyard Henry Julius W. - Wellesz Egon eds. (1935), Sticherarium, Copenhagen: Munksgaard (Mmb, Série Principale 1).

Jung Annette (1999), ‘The Long Melismas in the Non-kalophonic Sticherarion’, Cahiers de l’Institut du Moyen-Âge grec et latin 70, pp. 13-79.

Marchi Silvia (1996), I manoscritti della Biblioteca Capitolare di Verona. Catalogo descrittivo redatto da don Antonio Spagnolo, Verona: Mazziana.

Μηναῖα τοῦ ὅλου ἐνιαυτοῦ (Menaia di tutto l’anno) (1888-1901), 6 vols., Roma: Propaganda fide, vol. 6 (1901).

Mioni Elpidio (1965), Catalogo di manoscritti greci esistenti nelle biblioteche italiane, 2 vols., Roma: Istituto poligrafico dello Stato.

Papageorgios Spyridon (1907), ‘Τα ἐν Ἑνετίᾳ μουσικά χειρόγραφα’ [I manoscritti musicali conservati a Venezia], Φόρμιγξ Β΄, Β΄, pp. 23-24

Perria Lidia - Raasted Jörgen eds. (1992), Sticherarium Ambrosianum, Copenhagen: Munksgaard (Mmb, Série Principale 11).

Piazzi Alberto (1994), Biblioteca Capitolare, Verona, Fiesole: Nardini.

560 561 Gerda Wolfram

The Byzantine tradition of the great doxology

In Byzantine tradition the great doxology belongs to the office of matins. It is sung every Sunday, also at the great feasts of the Dodekaorton (i.e. the feasts of Jesus Christ and the feast days of the Theotokos), after lauds, after the psalms 148-150, as well as before the trisagion, which leads over to the Holy Liturgy. In the Latin west the great doxology, the Gloria in excelsis Deo, has been sung in the Holy Liturgy since the sixth century (Jungmann 1986, p. 356). The Apostolic constitutions are the earliest source of the Greek doxology from about 380 (Lodi 1979, n. 603, p. 373). The text, which was the main basis for the Latin tradition, has been handed down in the Codex Alexandrinus from the beginning of the fifth century (GB-Lbl Royal I.D.VII, fol. 569; Atkinson 1989, pp. 84-85). In Alfred Rahlfs’ Septuaginta edition, the great doxology is the fourteenth canticum, called «ὕμνος ἑωθινός» or morning hymn (Rahlfs 1935, 2, pp. 181-183). Rahlfs also uses the Codex Alexandrinus as his source for the great doxology. In this volume the Biblical cantica, the cantica of the New Testament and the great doxology are united. The first information about the great doxology we get from nun Egeria (Arce 1980), who was on pilgrimage in the Holy Land between the years 381-384. Her descriptions concern the cathedral offices of Jerusalem (Taft 1993, pp. 54-55). At matins after the morning psalms the Doxa en hypsistois was sung. Since the earliest times it might have been sung from beginning to end by the congregation. But also popular participation can be assumed by singing certain refrains, e.g. ἐλέησον ἡμᾶς. Egeria also informs us that in the second half of the fourth century antiphonal singing was already in use. Another important source for the early ages of Palestinian liturgy is the lectionary of the Georgian community, which translated the services of the urban rite of Jerusalem into its own language (Tarchnišvili 1959-1960; Leeb 1970, pp. 184-187). It is a redaction compiled between the fifth and the eighth century. Here we find information on the great doxology, sung at Holy Week and Easter Sunday in the following sequence: great doxology and Kataxioson (Καταξίωσον, Κύριε, καὶ τὴν ἡμέραν ταύτην ἀναμαρτήτους φυλαχθῆναι ἡμᾶς, verses 32 and 33 of the Greek doxology) poetic hymn, sung at the ascend of the clergy to the altar psalmody readings from the Holy Scripture poetic hymn trisagion

563 CANTUS PLANUS 2014 WOLFRAM — THE BYZANTINE TRADITION OF THE GREAT DOXOLOGY

Liturgical typika, which are at our disposal from about the tenth century onwards, with doxa-refrains are written in paleobyzantine adiastematic Chartres notation. only rarely contain information regarding the great doxology. References can be They belong either to the pentekostarion stichera, for the time between Easter and found only for particular feasts, e.g. the Elevation of the Cross (14 September), Pentecost, or to the stichera anastasima of the oktoechos, a cycle of Resurrection the Holy Week, and Easter. We can therefore assume that the melody of the great hymns in the eight modes. All of them are so-called ‘apokrypha’, i.e. hymns which doxology was much more elaborate on important feast-days than on lesser feasts. fell into disuse during the eleventh-twelfth century. The marginal notes to these One piece of information can be found in the liturgical typikon of the Great Church stichera are «εἰς τοὺς αἴνους» (sung at lauds) and «εἰς τὸ δόξα» respectively «εἰς τὸ δόξα of Constantinople, a manuscript from the tenth century. It tells us that the great ἐν ὑψίστοις» (sung during the great doxology). In Batopedi 1488 fourteen hymns close doxology is sung at matins after lauds before the trisagion (Mateos 1962-1963, 1, p. with the first part of the great doxology, more or less notated. 82). Unfortunately, no music-liturgical books of cathedral rite of the Hagia Sophia of Constantinople written in that period are preserved. The musical tradition was The incipits of the hymns and their modes: influenced by imperial and patriarchal ceremonies, predominantly texts from the Κύριε ὁ θεός μου, ὁ ἰασάμενος (πλ. β΄) psalter and a small number of non-Biblical hymns. We speak here of the akolouthia ( ) asmatike, the chanted office of Hagia Sophia, that was sung throughout (Lingas Ὁ Θωμὰς τῇ ψηλαφήσει πλ. β΄ 1996, pp. 100-110). Around the mid-ninth or early tenth century – with the rise Σὺν ταῖς ἄλλαις ὁλοφυρομένη (β΄) of neumatic notation – a rich melodic elaboration of certain chant repertories for Τὸ ὕδωρ τὸ ζῶν (πλ. β΄) the cathedral rite began: the so-called kontakia, florid but conventional formulaic Τὸ ἄντλημα καὶ τὸ φρέαρ (β΄) hymns, were created (Strunk 1977, p. 160). During the twelfth century, monastic and non-monastic practices started to intermingle in the Byzantine church. By 1204, Ἐπὶ τὸ ὄρος τῶν ἐλαιῶν (α΄) when crusaders conquered Constantinople, the rites of the Great Church came to an Μὴ θρήνοις ἐμποιεῖσθαι (πλ. δ΄) end (Strunk 1977, pp. 112-150). But very late, by the end of the twelfth or beginning Τῆς τελευταίας ἑορτῆς (πλ. β΄) of the thirteenth century, the musical tradition of Hagia Sophia was handed down by two chant-books in middle Byzantine diastematic notation; on the one hand, the The following hymns are staurotheotokia, which belong to the oktoechos: psaltikon for the soloist, containing kontakia and alleluiaria, on the other hand, ( ) the asmatikon with choral chants for the mass ordinary of the Constantinopolitan Παρεστώσα τῷ σταυρῷ σου Κύριε α΄ eucharistic liturgies. Parts of the ordinary chants of the cathedral tradition did Ὥσπερ τῆς παρθενίας (β΄) not appear before the fourteenth century, when the great μελῳδοί influenced the Παρθενικὴν ὥσπερ μήτραν (γ΄) musical genres with their new kalophonic style. Ne­vertheless, these new akolouthiai Τί σου τὸ ξένον υἱέ μου (δ΄) inherited some chants from the older repertory of the asmatikon and the psaltikon, as will be mentioned below. Ἐπὶ ξύλου βλέπουσα κρεμάμενον (πλ. β΄) Εἰ καὶ λίθος βαρύς (πλ. α΄) In contrast to the cathedral rite of Constantinople, incipits of notated doxologies can be found in sticheraria, books for the great choir with poetic hymns, which were sung Each hymn ends with a short phrase leading over to the doxology, as can be seen in only once a year, inserted between the psalm-verses. The repertory of the sticherarion these examples: originates in the monastic tradition of Palestine. Here the most famous melodes σὺν αὐτῷ βοῶμεν σοι (with him we call for you) such as John of Damascus, Kosmas of Maiouma or Andrew of Crete substantially (together with them we call for you incessantly) influenced the Byzantine tradition during the seventh and the eighth century. After μεθ᾿ ὧν ἀκαταπαύστως βοῶμεν Palestine were conquered by the Arabs, during the eighth century, the tradition σὺν πᾶσι τοῖς πιστοῖς κράζοντες (with all the faithful we cry) shifted to Constantinople, to the Studiou monastery, where a great amount of new διὸ κραυγάσατε πᾶντες (therefore all of you call out) hymns was added to the repertory. From the tenth century onwards the monasteries of Mount Athos too exerted great influence on yzantineB hymnology. The following In the manuscript A-Wn Theol. gr. 136 (henceforth Theol. gr. 136) from the beginning examples are taken from Codex Athos, GR-AOva 1488 (henceforth Batopedi 1488), of the twelfth century (Wolfram 1987), a sticherarion written in adiastematic Coislin dating from around the year 1050 (Follieri and Strunk 1975). All mentioned stichera notation, such doxa refrains can be found too. They are connected to Christmas hymns:

564 565 CANTUS PLANUS 2014 WOLFRAM — THE BYZANTINE TRADITION OF THE GREAT DOXOLOGY

Εὐφραινέσθωσαν οἱ οὐρανοὶ καὶ ἀγαλλιάσθω ἡ γῆ σήμερον (πλ. α΄) Example 1. Manuscript A-Wn Theol. gr. 185, fols. 109v-110 Σήμερον τίκτει ἡ παρθένος (γ΄) Μέγα καὶ παράδοξον θαῦμα τετέλεσται σήμερον παρθένος (β΄)

We can assume that the refrain Δόξα ἐν ὑψίστοις Θεῷ καὶ ἐπὶ γῆς εἰρήνη ἐν ἀνθρώποις εὐδοκία (sic) was sung by the choir. Whether the doxology was continued by a priest or by a soloist can only be an assumption based on the examples of the fourteenth century. It should be noted here that the doxa-refrain is always sung in the second mode respectively the second plagal mode, though the preceding sticheron can also belong to a different echos. The second mode has its fundamental tone in G, its upper initial tone in b, while the second plagal mode has its fundamental tone in E and its upper initial tone in G. It can be seen that the second authentic mode has another initial melody than the second plagal mode. In our examples from the Vatopedi manuscript we can see that in those cases where the sticheron belongs to a different mode, the signature for the second mode appears before the doxology. In the post- Byzantine period from the seventeenth century onwards, great doxologies were composed in all eight modes. There are no other notated examples to be found in the liturgical books between the eleventh and the fourteenth centuries. Only few mass ordinaries for the great feasts are recorded in musical manuscripts, for instance in the already mentioned Theol. gr. 136 from the beginning of the twelfth century, for In manuscript GR-An 2458, written in 1336, a great doxology can be found on fol. 123v Holy Saturday: εἰς τὸ Δόξα ἐν ὑψίστοις γίνεται εἴσοδος (at the doxa the entrance takes within an antiphon. It is written in the second plagal mode for the midnight service, in place; see Wolfram 1987, fol. 232). This means that the clergy goes to the altar and honour of the archangels. It comprises the great doxology until «ἐν ἀνθρώποις εὐδοκία», the choir begins to sing the trisagion. the trisagion and a small doxology. The melos is much simpler than the example from the Viennese manuscript. It has an ascending ambitus of E–d, while the melos in the In the manuscript A-Wn Theol. gr. 185 (henceforth Theol. gr. 185), an akolouthia Viennese manuscript moves upwards from G to g and downwards to D. manuscript from the last quarter of the fourteenth century, written in Thessaloniki, there is a notated great doxology, a reminiscence of the tradition of the chanted office Example 2. Manuscript GR-An 2458, fol. 123v of Hagia Sophia (see example 1). It is a manuscript where the late-Byzantine musical style of kalophonia is predominant. This doxology is characterised by long vocal ebellishments, as we can find them in the melodies of the kontakia. Here the singing of the doxology is divided between an anagnostes (the reader), a domestikos (the leader of the choir), and the priests (Lingas 1996, pp. 105-106). In Theol. gr. 185 the anagnostes enters the ambo with a cross and proclaims «Δόξα ἐν ὑψίστοις Θεῷ». The domestikos responds with a florid melos Δόξα« ἐν ὑψίστοις Θεῷ καὶ ἐπὶ γῆς εἰρήνη», with long vocalisations. The melos concludes with meaningless syllables. The insertion of syllables such as ου, γγω, χε, χι, να are characteristic for the psaltic genre. The priests We can see that there existed at least three different traditions: one in the second continue with «ἐν ἀνθρώποις εὐδοκία. ὑμνοῦμεν σε, εὐλογοῦμεν σε, προσκυνοῦμεν σε», mode, a simpler one in the second plagal mode, and one in the sticheraric tradition. at «εὐχαρι […]» it breaks off. It is a syllabic chant for the priests of the great doxology We can assume that during the ninth-tenth century, there already existed a (Lingas 1996, pp. 104-107). The melos of the priests moves upwards and downwards melismatic doxology for great feasts in the Hagia Sophia tradition. But we have no within a fifth in a very simple way. We can assume therefore that the performance evidence whether the whole chant was sung by a soloist or whether it was continued of the remaining­ verses continued in the same manner. Instead of «αἰνοῦμεν σε» the by the priests at that time. From the Viennese example we can conclude that the priests sing «ὑμνοῦμεν σε». soloist performed only the first part of the oxologyd with a very melismatic passage

566 567 CANTUS PLANUS 2014 and the priests continued in syllabic style. Michel Huglo, our dear late colleague, Nina-Maria Wanek made an edition of the text and the melody of the doxology, based on manuscripts from Saint-Martial and Laon, to reconstruct the Greek origin of the hymn (1950, pp. O quando in cruce – Ὅτε τῷ σταυρῷ revisited 30-40). At the beginning the melody on G-a-b corresponds to the psaltic version in the second mode, as can be seen in example 1, but the formulas are different to those represented in the Viennese manuscript. Perhaps the Latin version hands down the musical style of the asmatikon or a melismatic version of the monastic tradition. Since we know only few Greek examples of the great doxology it is difficult to draw conclusions from the available material. The troparion O quando in cruce (Ὅτε τῷ σταυρῷ) is one of several bilingual Latin- Greek antiphons for the Adoration of the Cross which can be found in Beneventan and Ravenna sources dating from the eleventh-twelfth century. It is especially appropriate for a direct comparison with extant Byzantine and western versions, because, contrary to the chants of the Missa graeca, there is no time-gap of several centuries: the troparion is included in nearly contemporaneous manuscripts in the east and west. Bibliography The Byzantine melody can be found in manuscripts from the tenth century onwards,

Arce Agustin ed. (1980), Itinerario de la virgen Egeria (381-384), Madrid: Biblioteca de Autores Cristianos. the western codices containing O quando in cruce date from the eleventh century. Egon Wellesz was one of the very first scholars to draw attention to the fact that the melody Atkinson Charles M. (1989), ‘The «Doxa», the «Pisteuo», and the «ellinici fratres»: Some Anomalies in the of O quando in cruce corresponds to the Byzantine troparion . In 1947 Transmission of the Chants of the «Missa Graeca»’, Journal of Musicology 7, pp. 81-106. Ὅτε τῷ σταυρῷ he presented his comparisons and results in his book Eastern Elements in Western Follieri Enrica - Strunk Oliver eds. (1975), Triodium Athoum, Copenhagen: Munksgaard (Mmb, Série Chant. In the seventy-three years since Wellesz published this book, O quando in cruce Principale 9). has hardly ever been investigated again, one exception being Thomas F. Kelly (1989), Huglo Michel (1950), ‘La Mélodie Grecque du «Gloria in Excelsis» et son utilisation dans le Gloria XIV’, who mentions the troparion in his book on Beneventan Chant. The present article1 Révue Grégorienne 29, pp. 30-40. will concentrate on the transliterated text of the troparion in western manuscripts,

Jungmann Josef A. (1986), The Mass of the Roman Rite, its Origins and Development, Dublin: Ave Maria. as it is still too early to thoroughly assess the differences respectively similarities of the melodies in the east and west. The Byzantine chant Ὅτε τῷ σταυρῷ belongs to Leeb Helmut (1970), Die Gesänge im Gemeindegottesdienst von Jerusalem (vom 5. bis 8. Jahrhundert), Wien: Herder (Wiener Beiträge zur Theologie 28). the genre of the so-called troparia, meaning a «strophic poetic chant that reoccurs between verses of a psalm […] in this case it refers specifically to the final troparion Lingas Alexander L. (1996), Sunday Matins in the Byzantine Cathedral Rite: Music and Liturgy, Ph. D. of each office after the trisagion» (Touliatos-Miles 2010). Wellesz (1961, p. 171) writes: dissertation, University of British Columbia, Vancouver. The name troparion […] was given to short prayers which, in the earliest stage of hymnography, Lodi Enzo ed. (1979), Enchiridion euchologicum fontium liturgicorum, Roma: Clv. Edizioni liturgiche. were written in poetic prose and inserted after each verse of a psalm. In the fifth century, when the Mateos Javier (1962-1963), Le Typicon de la Grande Eglise. Ms. Sainte-Croix N° 40, 2 vols., Roma: Pontificio Troparia were composed in strophic form and became longer, these poetical prayers were sung Istituto Orientale (Oriens Christianus Analecta 165-166). only after the three to six last verses of a psalm. Hymns of this kind are known to have formed part of matins and vespers in churches and monasteries of the fifth century. Rahlfs Alfred (1935), Septuaginta, 2 vols., Stuttgart: Deutsche Bibelstiftung.

Strunk Oliver (1977), Essays on Music in the Byzantine World, New York: Norton. The Latin term tropus (which also signifies a short strophe) is strikingly similar to the Greek word tropos, a fact already pointed out by Egon Wellesz (1947, p. 155). As Taft Robert (1993), The Liturgy of the Hours in East and West. The Origin of the Divine Office and its Meaning for Today, Collegeville: The Liturgical Press. Wellesz stated, Ὅτε τῷ σταυρῷ might be attributed to saint Sophronios, patriarch of Jerusalem (634-638), according to a manuscript from the year 1122 containing the e e Tarchnišvili Michel ed. (1959-1960), Le grand Lectionnaire de l’église de Jérusalem (V -VIII siècle), 4 typikon of the Church of Jerusalem.2 The troparion has its place among the chants of vols., Louvain: Sécretariat du Corpus Scriptorum Christianorum Orientalium (Corpus Scriptorum Christianorum Orientalium 188-189, 204-205 = Scriptores Iberici 9-10, 13-14). 1. The present paper is part of the research-project P27115 funded by the Austrian Science Fund (Fwf).

Wolfram Gerda ed. (1987), Sticherarium antiquum Vindobonense. Codex Theol. gr. 136 Bibliothecae 2. Wellesz (1947, p. 22 ) writes: «The antiphons sung during the ‘Holy Hours’ are given in full in the typikon of 1122 and in Nationalis Austriacae phototypice depictus, 2 vols., Wien: Verlag der österreichischen Akademie der the actual triodion; they are twelve short chants, each consisting of a single strophe […]». Number 8 of these troparia, which Wissenschaften (Mmb, Série Principale 10). are all attributed to patriarch Sophronios, is Ὅτε τῷ σταυρῷ.

568 569 CANTUS PLANUS 2014 WANEK — O QUANDO IN CRUCE – ὍΤΕ Τῷ ΣΤΑΥΡῷ REVISITED the sticherarion, specifically among the stichera sung at Good Friday at the ninth hour, Table 1. Text of the troparion as stated in the typikon of the Church of Jerusalem in the ninth or tenth century. It is preceded by the stichera Σῶσόν με, Κύριε, ὅτι ἐκλέλοιπεν ὅσιος and Μάταια ἐλάλησεν Greek MOd O.I.7 BV 38 Latin English ἕκαστος πρὸς τὸν πλησίον (Wellesz 1947, p. 85). Ὅτε/Ὅταν Ote ton stabron Otin to stauron O quando in cruce When to the cross τῷ σταυρῷ Greek and western manuscripts containing the troparion. Ὅτε τῷ σταυρῷ is προσήλωσαν proschiloson prosilosan confixerant iniqui transgressors included in a great number of sticheraria. For his study Wellesz used the manuscript παράνομοι τὸν paranomi ton paranomi, ton Dominum gloriae! nailed the Lord Κύριον τῆς δόξης, Kirion tis doxis Kyrion tis doxis, of Glory Athos, Μονὴ Μεγίστης Λαύρας GR-AOml 252 (fol. 45; henceforth AOml 252) from the aevoa pros athus evoa pros aptus: Ait ad eos: He cried out to them: ninth-tenth century. This is one of the earliest existing manuscripts containing the ἐβόα πρὸς αὐτούς· troparion. As its notation is still adiastematic an exact transcription is not possible, but Τί ὑμᾶς ἐλύπησα; thymas ydiclisas Ti yma ydikisan? Quid vobis Wherein have I molestus sum? grieved you the melodic line is recognizable enough in order to be compared to later examples. ἢ ἐν τίνι παρώργισα; gen tinis i en tini parogiso? aut in quo or wherein have Ὅτε τῷ σταυρῷ is also included in the manuscript A-Wn Theol. gr. 136, from the parorchisan iratus sum? I provoked your beginning of the twelfth century, fol. 228v (henceforth Theol. gr. 136). This manuscript wrath? is written in the so-called Coislin v notation, which is already diastematic enough to πρὸ ἐμοῦ pro emutis Pro emu Ante me Before me make a rough transcription into the western notation possible. The manuscript A-Wn τίς ἐῤῥύσατο ὑμᾶς ymas erisato tis imas erisate quis vos liberavit who delivered you Theol. gr. 181 (henceforth Theol. gr. 181) from the year 1221, contains the troparion ἐκ θλίψεως; e glipseos ec klipseos? ex angustis? from oppression? in middle Byzantine ii notation, too (fol. 249v). Wellesz also used as one of his main καὶ νῦν Ke nin Ke ni Et nunc And now sources the Codex Athos, Μονὴ Βατοπεδίου GR-AOva 1499 (fol. 297) from the year 1292 τί μοι ἀνταποδίδοτε; ti mi antapodidote ti mi antapodydote quid mihi redditis what return (henceforth Batopedi 1499). The manuscript I-Ma A 139 sup. from 1341 (henceforth do you make to me? A 139) is also a reliable source, and similar to Theol. gr. 181 and Batopedi 1499.3 Only πονηρὰ ἀντὶ ἀγαθῶν· ponira anti agatho ponira anti agotho? mala pro bonis? Evil for goodness. two western manuscripts are known, which not only contain the text in Greek, but Ἀντὶ στύλου πυρός, anti stilu piros Anti stilu piros Pro columna ignis In return also have neumes added. Both are graduals from the eleventh century (Wellesz 1947, p. for a pillar of fire 68 and Plates ii-iii): I-BV 38, mature period of Beneventan scripts; I-MOd O.I.7, central σταυρῷ μέ stauron me stauron mu in cruce you have nailed me Italian type of the northern zones, origin: Ravenna (henceforth MOd O.I.7). προσηλώσατε· proskilosate prosilonsate; me configitis; to a cross. Ἀντὶ νεφέλης, anti nefelis anti nefeli pro nube In return for a cloud Wellesz (ibidem p. 69) after Joseph Gajard, who did the transcription for him, states: τάφον μοι ὠρύξατε· tapon mi orixate tafon mi erisate; sepulchrum you have dug me […] that the codex in the Library of Modena might have had its origin in Ravenna seems very mihi fodistis; a tomb. convincing to me, as Ravenna was one of the places through which Byzantine and Syrian Ἀντὶ τοῦ μάννα, anti tu manna anti tu manna pro manna In return ecclesiastical art and liturgical customs entered Italy. for the manna χολήν μοι choloim cholin mu fel me potastis; you have offered Other six manuscripts (I-BV 35, 39, and 40; V-CVbav Barb. lat. 603; I-Lc 606, and I-PCd προσηνέγκατε· prosinonchate epotisate; me gall. 65) have neumes added, but only to the Latin text. As Wellesz (ibidem, p. 69) found Ἀντὶ τοῦ ὕδατος, anti tu idatos anti tu ydato propter aquas In return for water out comparing theses codices, they all «correspond with each other, except Codex ὄξος με ἐποτίσατε· hocos mi epotisate osos mi epotisate? acetum mihi in you have given me MOd O.I.7 and also with the neumes of the Greek text in Codex BV 38, apart from poculum porrigitis. vine­gar to drink. insignificant differences due to the text’s being in another language». Kelly (1989, p. Λοιπὸν καλῶ lypon Kala Lipon kala Ergo vocabo For the future 207) writes about the occurrence of the troparion in Beneventan manuscripts that τὰ ἔθνη ta etni ta etni gentes I will call the Gentiles the bilingual antiphon O quando in cruce, though it is transmitted in southern Italy is, like the κἀκεῖνα με cakine me doxasusi kekyna me ut ipsi me and they shall trisagion, not a part of the older corpus of the Beneventan liturgy. It is neither included in our δοξάσουσι doxasusin glorificent glorify me oldest sources [V-CVbav.] Vat. 10673 and BV. 33, nor mentioned in the ordo of Vat. 10673, and the σὺν Πατρὶ sin Patri syn Patri una cum Patre with the Father variety of its placement in Beneventan­ manuscripts suggests its relatively recent arrival. καὶ ἁγίῳ Ke agion kai Agio et cum and the Pneumatin. Pneumati. Sancto Spiritu. Holy Spirit.* 3. There are a great number of manuscripts from the National Library of Greece in Athens, which also contain the Πνεύματι. troparion (e.g. GR-An 889, 890, and 892 from the thirteenth century, 883 and 884 both from the fourteenth century, or 885 from the fifteenth century). * The English text is taken from Wellesz (1947, p. 23).

570 571 CANTUS PLANUS 2014 WANEK — O QUANDO IN CRUCE – ὍΤΕ Τῷ ΣΤΑΥΡῷ REVISITED

Comparison of the transliterated text 38 always uses the form «stauron». MOd O.I.7’s usage of «stabron» clearly shows Kelly (1989, p. 217) states in his book that that the diphthong here was already pronounced as «af» like in Modern Greek. One should not forget though that even if the form «stauron» can be found in the Latin [… the] transliteration [of Panta ta etni [sic]/Omnes gentes, Ton stauron/Crucem tuam and Otin to manuscripts, the letter «u» was normally also used for «v» in Latin writing! «Οι» is stauron/O quando in cruce] in Latin characters shows the absence of a written Greek tradition always «i» or «y» in MOd O.1.7 («lypon»), only once in BV 38 it becomes «u»: χολήν μοι in these manuscripts, and variations in spelling suggest writing from oral memory. The Greek is transliterated as «cholin mu», probably due to mishearing. is written down as it is heard: it comes from an oral source, and hence presumably­ one nearby. Krasis is a very common mistake in all transliterated texts. In the troparion the I would like here to continue the assessment of the transliteration which I have begun following examples can be found: ἐμοῦ τίς → «emutis» (MOd O.I. 7); ἐκ θλίψεως → «e at the 2012 Cantus Planus conference with the example of the bilingual Sanctus/Ἅγιος.4 glipseos» (MOd O.I.7) and «ec klipseos» (BV 38); χολήν μοι → «choloim» (MOd O.I.7); The transliterations and the mistakes in the troparion are very similar to those of the ἢ ἐν τίνι → «gen tinis» (MOd O.I.7). chants of the Missa graeca: Vowels. «O» is often mistaken for «a» due to the very open pronunciation of «o» in Spiritus asper. There are no aspirated vowels, e.g. for hotan/hotin/hote for Ὅτε/ Greek: e.g. ἀγαθῶν «agotho» (BV 38); καλῶ → «kala» in both manuscripts. Ὅταν Ὅταν or ὑμᾶς. becomes «otin» in BV 38. In the Byzantine version only two forms are used: Ὅτε (from ὅτε ἄν) and ὅταν, which both mean «when» and which is correctly translated «γι»: As in the Sanctus, this syllable here is usually rendered as «gi», e.g. ἁγίῳ at with «quando» in Latin. The form otin, though, is not known in Greek – neither in the end is in both codices transcribed as «agio(n)»; in the Sanctus we also found ancient, nor Byzantine nor in demotic Greek. We can only guess as to how the scribe «gy», and «y». In the troparion the word παρώργισα is transliterated differently in transliterated with «otin»: usually «a» is confused with «o» or vice versa. But both codices, namely «parorchisan» and «parogiso». «Parorchisa» in MOd O.I.7 is ὅταν perhaps here the scribe used the familiar-sounding «in» from «in cruce» and thus especially interesting, because it really tries to reproduce the sound «γι». In Greek came up with «otin». Interestingly enough this erroneous form was kept in the title «γι» is pronounced like the English «y» in «yes» and not as «g» as in «goose». The of the chant throughout the centuries.5 sound «y» in «γι» though is not as clear as in «yes», but slightly aspirated, giving the listener the impression that he hears a «ch» after the «y». If pronounced quickly, the «Υ» is mostly transcribed as «y», sometimes «i»; this is also the case in the «Hagios». syllables ώργι in παρώργισα indeed sound like «orch», the «y»-sound of «γι» being In the troparion we find «Kirion» (MOd O.I.7) and «Kyrion» (BV 38).Τί ὑμᾶς becomes swallowed up by the «r» before. This point is important, because the pronunciation «thymas» (MOd O.I.7) and «tyma» (BV 38). Nῦν is symply transcribed as «ni(n)» and of ώργι is rarely rendered correctly by someone whose mother-tongue is not Greek! ἐῤῥύσατο as «erisato/erisate»; ὑμᾶς is both «ymas» and «imas»; στύλου is «stilu» in both manuscripts, ὠρύξατε becomes «orixate/erisate»; ὕδατος «idatos» respectively Codex MOd O.I.7 also makes παρώργισα (first person singular) into fourth person «ydato» and σὺν is «sin/syn». BV 38 is the manuscript, which more often uses «y» for plural here, perhaps analogous to προσήλωσαν from the beginning of the chant. «υ». It also takes «y» for «ι», e.g. in «antapodydote». Interestingly the syllable σήλ of προσήλωσαν is in MOd O.I.7 again transliterated with an inserted «ch» («proschilosan»), probably again in an attempt to reproduce the «η» is always transliterated as «i» (as in the Sanctus too). exact sound the speaker or singer was making here. The letter σ in Greek is often Consonants. «Φ», e.g. τάφον is written «tapon» in MOd O.I.7 and correctly «tafon» in pronounced slightly aspirated, making a sound like «s-ch» contrary to the letter , ζ BV 38. which is a voiced «z» as in «zoo». MOd O.I.7 transliterates the second προσηλώσατε as «proskilosate». BV 38 writes «parogiso», making it clear that here the scribe did «β», e.g. ἐβόα is always transcribed as «v» («aevoa» and «evoa»), never as «b». not hear the «r» of the syllables ώργι, obviously because the «ρ» here tends to be «ξ» in ὄξος seems to have caused difficulties, although the sound as well as the letter ‘swallowed’ up, making the «ώ» even more ‘open-sounding’ than it normally is. «x» are common in Latin. Perhaps the combination «oxo» was strange for a non- Diphthongs. «οὐ» is always «ou» (in the Sanctus some manuscripts used indeed «ou»). Greek: in MOd O.I.7 it is transliterated as «hocos», although there is no spiritus asper at the «o» and «osos» in BV 38. «αυ» in σταυρός is transliterated differently also in one and the same codex: MOd O.I.7 writes at the beginning «stabron» and the second time «stauron»; it always puts «θ» is always «t» («etni»), also in the other chant for Good Friday «Panta ta etni». an additional «n» at the end, although it is dative in Greek and not accusative. BV

5. I am indebted here to Andreas Rhoby (Austrian Academy of Sciences) for his information on the etymology and the 4. The analysis of the bilingual Sanctus/Ἅγιος can be found in Wanek (2013, pp. 175-192). occurrence of «ὅταν» throughout the centuries and his suggestion why the scribe might have made this error.

572 573 CANTUS PLANUS 2014 WANEK — O QUANDO IN CRUCE – ὍΤΕ Τῷ ΣΤΑΥΡῷ REVISITED

Regarding the above observations one gets the impression that the troparion in MOd As regards the comparison with the Byzantine melody in the diastematic codices of the O.I.7 was definitely written down by someone who did not have any knowledge of thirteenth century, Wellesz (1947, pp. 109-110) states that «the Western and Byzantine Greek at all, but nevertheless tried to give a very realistic reproduction of the sound melodies resemble each other enough in order to say that they have a common source». of the Greek words. Even more important, the scribe of MOd O.I.7 seems to have Kelly (1989, p. 217 and 218) writes that had a native Greek to read the text to him, because the pronunciation seems very authentic to spoken Greek, according to the spelling. Someone who has Greek not as at least two of these pieces (O quando in cruce and Crucem tuam adoramus) have parallels in the his mother-tongue usually pronounces some of the above-mentioned examples more Byzantine liturgy, but they were undoubtedly adopted at Benevento, not directly from the East, but through the intermediary of some source relatively near at hand where such music could be ‘clearly’ (see the aspirated « ») or pronounces the letters as he is used to in his own σ heard. The other pieces in Greek may also have been features of an Italian Greek liturgy. […] That mother-tongue. As Thomas Kelly (1989, p. 218) states, «Greek influence was always the adorations appear also in manuscripts of Ravenna suggests that they derive from an ancient present at Benevento. The Byzantine garrison there at the end of the ninth century Italian practice, perhaps independent of Rome, which underwent subsequent development in brought a substantial Greek population of officials, soldiers, and the necessary clergy various ways, the Beneventan and Ambrosian versions retaining a close relationship from which for the Byzantine rite». It might not have been uncommon then to find someone, most the Ravenna practice, at least as it is preserved now, varies substantially. probably among the clergy, whose mother-tongue was Greek to read the relevant text to the scribes. Kelly goes on to explain: The thirteenth-century manuscripts and even Theol. gr. 136 show a much more developed melody than the western ones, which are mainly syllabic except for a Despite the hostility of the inhabitants, churchmen would be aware of one another’s presence, and few melismas on the words e.g. ἀνταποδίδοτε, μάννα and πατρί. The best source for of the difference in their rites; and though the Greeks would hardly borrow materials from their comparing the melodies would be the athonite Codex AOml 252), which is as syllabic subjects, the long tradition of Greek in the Western church at large, and in the Beneventan church as MOd O.I.7 and BV 38. As was stated above, AOml 252 is still adiastematic and does in particular, might lead to the adoption of further Greek items (ibidem, p. 218). not even apply neumes to all words or syllables, so comparison will be very difficult. This is the point were, in due course, I will be hopefully able to work in more detail on Codex BV 38 shows a more ‘common’ transliteration, perhaps done by someone comparison between the western and Byzantine manuscripts from the tenth-eleventh who had a rudimentary knowledge of Greek, as we can find it in many manuscripts century. Because of the adiastemacy of the latter, it is very hard to come to a conclusion containing parts of the Missa graeca and other bilingual chants. regarding differences respectively similarities between eastern and western melodies here. As far as I have seen until now, there seem to be quite a few differences between Byzantine and western melodies AOml 252 and MOd O.I.7 respectively BV 38. The new project (see note 1) will attempt Regarding the melodies, it would need a separate study to make a clear statement. a reconstruction of the melody in AOml 252 to lay the basis for a detailed comparison The above-mentioned Byzantine manuscripts Batopedi 1499 and its contemporary with the western manuscripts. Theol. gr. 181 clearly contain the same melody with only the slightest variations as they are common in many Byzantine chants, namely exchanging the leap of a second with a third, an ison with a second etc., without altering the melody significantly (Wellesz 1947, Plates ix-x). The fourteenth-century manuscript A 139 also shows the same melody as in the earlier codices with only minor alterations. The melodies of the twelfth-century codices, e.g. the already mentioned Theol. gr. 136, but also I-GRe Δ. β. X (ca. 1138), show hardly any differences when compared to each other and seem to be the basis for the later codices, although there are differences in the melodic outline and they contain of course a less elaborate version omitting melismas which the later codices have added (Wellesz 1947, Plates ix-x). Also the melody in codex AOml 252 bears resemblances to the twelfth-century codices, but being still adiastematic it will need more time yet to reconstruct the melody sufficiently for a thorough comparison. The western melodies in MOd O.I.7 and BV 38 (see example 1, p. 576) show great resemblances among each other and it is safe to say that the melody is the same in both codices, while BV 38 has more melismas and in a more developed state.

574 575 2 CANTUS PLANUS 2014 WANEK — O QUANDO IN CRUCE – ὍΤΕ Τῷ ΣΤΑΥΡῷ REVISITED

Wanek es. 1 Ote to stauro Example 1. Melodic line, providedWanek with Greek es. text1 Ote (without to stauro breathings and accents) Comparison between Latin sources. 1 = BV 38, fol. 42; 2 = MOd O.I.7, fol. 99

1 œ 1 œ œ œ & œ œ œ œ œœ œ œ œ œ œ œœ œœ & œ œ œ œ œœœ œ œ œ œ œ œ œ œ Οταν τω σταυ− ρω προσ− η −− λω σαν πα − ρα − νο −μοι τον προσ−− η λω − σα − τε αν − τι νε − ϕε − λης τα − ϕον μοι ω − ρυ − ξα − τε 2 œ 2 œœ œ & œ œ œ œ œœ œ œ œ œ œ œ œœ œ œ & œ œ œ œ œ œœœ œ œ œ œ œ œ œ œ œ œ

1 1 œ œ œ œœ œ œ œ œ œ œœœ œ œ œ œ & œœœ œ œ œ œ œ œœœ & κυ−− ριœ ον της δο− ξης ε − βο − α προς αυ − τους τι υ− μας ε − αν− τι του μα− να χο− λην μοι προσ−− η

2 œ œ œ œ 2 œ œ œ œ œ & œ œ œ œ œ œœ œ œ œœœ œ œ œ œ & œœœ œ œ œ œ œœœ œ œ œ œœœ œ œ

1 1 œ œ œ œ œ œ œœœ œ œ œ œœ œ œœ & œ œ œ œœ œ œ œ œ œ œ œ œ œ œœœ œ & λυ−− πη σαœ η εν τιœ − νι παρ − ωρ − γι − σα; œœπρο ε− μου νεγ−− κα τε; αν − τι του υ − δα − τος

2 2 œ œ œ œ œ & œ œ œ œ œ œ œ œ œœœ œ œ œ œœœ œ œ œœ & œ œ œ œœ œ œ œ œ œ œœœ œ

1 1 œ œ œ œœœ œ œ œ œœœœ œ œ œ œ & œœ œ œ œ œ œ œ œ œ œ œ œ œœ œœœ œ & τις ερ− ρυ −− σα το υ − μαςœœ εκ θλι − ψεœ − ως; και νυν τι μοι αντ− α ο− ξος με ε πο−− τι σα − τε; Λοι − πον κα − λω τα ε − θνη κα − κει − να

2 œ œ 2 œ œ & œœœœœ œ œ œ œ œ œœœ œ œ œ & œ œ œœœ œ œ œ œ œ œ œ œ œ œ œ œ œ œ œ

1 œ 1 œ œ œ œ œ œ œ œ œ œ œœ œ œ œ œ œ & œœœ œ œ œ œœ œ œ œ œ œ œ œ œ œ œœ œ œ & πο−− δι δο − τε œ ποœ − νη ρα− αν − με δο− ξα −− σου σι συν πα − τρι

2 œ œ œ œ œ 2 œ œ œ œ œ œ œ & œ œ œ œ œ œœ œ œ œ œ œ œ œ œ & œœœ œ œ œ œ œ œ œ œ œœ œ œ

1 1 œ œ œ œ œ & œ œ œ œ œœœœ œ œ œ œ œ œ œœ & œ œ œ œ œœ œ œ œ τι α−− γα θων; αν − τι στυ − λου πυ − ρος σταυ− ρω με και α −γι − ω πνευ−− μα τι.

2 œ 2 & œ œ œ œ œœœœ œ œ œ œœœ œ œœ & œœœ œ œ œ œ œ œ œ œ œ œ

576 577 1 I-MOd 2 I-BV 38 CANTUS PLANUS 2014

Svetlana Poliakova Bibliography

Kelly Thomas F. (1989), The Beneventan Chant, Cambridge: Cambridge University Press. The classification of Russian sticheraria: – (2011), The Sources of Beneventan Chant, Farnham: Ashgate. a preliminary case study

Touliatos-Miles Diane H. (2010), A Descriptive Catalogue of the Musical Manuscript Collection of the National Library of Greece. Byzantine Chant and Other Music Repertory Recovered, Aldershot: Ashgate.

Wanek Nina-Maria (2013), ‘Die sogenannte Missa graeca. Schnittstelle zwischen Ost und West?’, Byzantinische Zeitschrift 106, pp. 175-192. This paper1 concerns all Russian neumatic sticheraria of the triodion cycle (collections of stichera from the weeks preparatory to Lent to the Sunday of All Saints) of the earliest Wellesz Egon (1947), Eastern Elements in Western Chant, Oxford: Oxford University Press. period of Russian liturgical chant, written between the twelfth and the thirteenth – (1961), A History of Byzantine Music and Hymnography. Oxford: Oxford University Press. centuries. Sticheraria older than the twelfth century did not survive or did not exist, though the use of Slavonic books in Russia can be traced to the end of the ninth-early tenth centuries. After the thirteenth century the production of sticheraria paused, as well as the production of Russian neumatic books in general, until the moment when at the turn of the fifteenth-sixteenth centuries a new phase of worship, book production, and system of notation began, and a new generation of Russian sticheraria appeared.

There survive seven ancient Russian sticheraria of the triodion cycle and a set of triodion and pentekostarion from the collections of the Historical Museum in Moscow (henceforth GIM), Synodal Collection (henceforth Synod.) 319, and Voskresensky Monastery collection (henceforth Voskr.) 27, including the notated part of the sticherarion in its complete form.2 These sources were studied to a varying extent.3 However, a complete picture of the oldest Russian sticheraria as a group with unifying and dividing characteristics, which is in some way part of the tradition of the Byzantine sticherarion, does not yet exist.

This paper aims to consider some unifying and dividing characteristics of the oldest Russian sticheraria, providing a starting point for further comparative study of thesesources. The Russian sticheraria to be discussed here are: (a) sticherarion Synod. 278, second half of the twelfth century,4 possibly of Novgorodian origin; (b) Chilandar sticherarion 307 (see note 3; henceforth Chil. 307), also dated to the twelfth century, and also considered to be from the Novgorod area; (c) sticherarion Typ 147 = Russian State Archive of Ancient Acts in Moscow, Synodal Typography Collection 381

1. This research was supported by Cesem at the Faculdade de Ciências Sociais e Humanas, Universidade Nova de Lisboa, under project UID/EAT/00693/2013, funded by the Fundação para a Ciência e Tecnologia, Portugal.

2. Due to scarcity of the content, a small fragment of pentekostarion of the twelfth-thirteenth century preserved in St. Petersburg, Russian National Library, Pogodin Collection 46, containing four notated folios of the Pentecostal celebration, was not regarded in the present article.

3. One of the manuscripts was published and studied in some detail. The main part, preserved in the Chilandari Monastery of Mount Athos (GR-AO Chil. 307), was published in the series Mmb (Jakobson 1957); two fragments of one folio each are preserved in St. Petersburg (BAN, Dmitr 44, RNB Q.п.I.39), published in Schidlovsky (2002); three folios are conserved at the National Museum in Prague (Mares 1978). The GIM triodion set was the subject of my Ph. D. dissertation; for the history of the studies of these manuscripts refer to Poliakova (2009).

4. There exist some disagreements concerning the dating and place of writing of some of the manuscripts (Poliakova 2009, pp. 146-149).

578 579 CANTUS PLANUS 2014 POLIAKOVA — THE CLASSIFICATION OF RUSSIAN STICHERARIA

(henceforth Typ.) no. 147; it dates from the twelfth century, or the beginning of the all the three cycles (from the menaion, triodion-pentekostarion, and oktoechos) may thirteenth; the possibility of Novgorodian origin cannot be excluded; (d) sticherarion be included in a single codex. There also exist sticheraria with the sequences of Typ. 148 also suggests the end of the twelfth century; apparently of Novgorodian origin, exclusively triodion-pentekostarion cycle. but the possibility of it having been written in Novgorod for Pskov is not excluded; (e) sticherarion Sof. 96 = Russian National Library of St. Petersburg, Collection of the All old Russian sticheraria up to the fifteenth century include only one annual cycle, Sofisky Cathedral of Novgorod (henceforth Sof.) 96; it probably dates between the and they are either sticheraria of the triodion and pentekostarion or sticheraria of twelfth and the fourteenth century, the first half of the thirteenth century being the the menaion.5 Thus, the detaching of the sticherarion of the triodion cycle, to form a most likely date; the fact that it belonged to the library of the Cathedral of St. Sophia separate codex observed in all seven sources (not included here are the multi-genre in Novgorod may suggest Novgorodian origin; (f) sticherarion Sof. 85, whose dating triodion and pentekostarion from the GIM collection) is a unifying characteristic of old and provenance are far more certain, was written in Novgorod between 1224 and Russian sticheraria setting them apart from the Byzantine tradition. 1226; (g) sticherarion Usp. 8 = GIM, Collection of the Cathedral of the Dormition of the Kremlin in Moscow (henceforth Usp.) 8; belonging to the thirteenth century, If we consider the commemorations in Russian sticheraria and the triodion set it is distinguished from other sticheraria of this sort by its distance from sources of from the Historical Museum collection, we note that the system of commemorations Novgorodian tradition (Металлов 1906, p. 209; cited in Schidlovsky 2002, p. 114); it is generally corresponds to the Studite type of typikon; most of the Constantinopolitan possible that it was written in Rostov: in the margins, in a fifteenth century hand the commemorations are present in them. At the same time, on the Sundays of Lent all stichera of the Menaion for Saint Leonty of Rostov were added (fol. 198v). So, from the manuscripts contain the commemorations which were transmitted from the Jerusalem point of view of the date of writing the sticheraria and the triodion set belong to the Gospel readings to the Studite synthesis – the Prodigal Son and Repentance for the same period, to its mature stage. From the point of view of local origins the group of Second Sunday, the Publican and the Pharisee for the Third Sunday, The One who these manuscripts can be divided into two unequal parts, i.e. those closed to Novgorod fell amongst Thieves for the Fourth Sunday, and the Rich Man and Lazarus for the and one of a distinct origin. Fifth Sunday. The characteristic which seems to be peculiar to the Russian sticheraria and which distinguishes them as a group is the regular fixing on the Second, Third, The notating style represent some unifying characteristics: all the sticheraria have Fourth and Fifth Sundays of the themes of Jerusalem commemorations in the titles.6 znamenny notation, which joins the marks of two different stages: according to Strunk Converging in the overall composition of the commemorations, the manuscripts differ (1977, p. 222), an archaic Coislin variety introduced into Slavic practice in around in assigning the Adoration of the Cross to the third Sunday of Lent (Usp. 8) or the 950 was revised in the middle of the eleventh century. At the same time, each of following Wednesday (most of the manuscripts), the Hours with the twelve troparia to the manuscripts represents notational variants, more significant for the melismatic Holy Friday (Voskr. 27, Typ. 147, Typ. 148, Sof. 85, Chil. 307; in Synod. 278 this sequence passages in some idiomela; the origins of these deviations are still unclear. is missing) or Holy Tuesday (Sof. 96 and Usp. 8).7

The textual redaction of the oldest Russian sticheraria seems to have been marked by As for the numbering of the weeks in the triodion and pentekostarion, Russian sources the similar periodization scheme. The mid tenth century is the epoch when Christianity are characterized by two systems for numbering the Lent period.8 In the former, the was introduced in the Russian state, culminated with the official ‘mass baptism of Rus’ weeks are counted chronologically, beginning with the First Sunday (dedicated to in 988, and the circulation of Greek books and already translated Slavonic books in the Prophets and to Orthodoxy), up to the Sixth Sunday, which is not indicated by a Russia could be supposed for this period. The process of the formation of Russian number but with the name, i.e. Palm Sunday. The Synod. 330, a Russian copy of the sticherarion and its connection with the Greek tradition and especially the South Slavic Studite typikon, which matches in a great degree with the sticheraria and the GIM tradition is not still clear. The later historical level testified by the notation in Russian triodion set due to a group of characteristics, follows the latter as well as triodion sticheraria coincides with the redaction of the liturgical books carried out in a single Synod. 319 and sticherarion Typ. 147. Russian centre in the second half of the eleventh century (Момина 1992, p. 215); in the course of this process new types of Russian books including the sticherarion came into existence. 5. Stichera of the oktoechos in Russian versions were normally collected with the kathismata in the Izborny oktoechos.

6. The only exception, on the Fourth Sunday in Typ. 147, may be due to an unintentional omission by the scribe. This Let us now consider the sticheraria from the point of view of the content of a codex. consistency in the titles suggests the existence of a common source close to the Jerusalem tradition. Ancient Byzantine sticheraria show a different kind of organization. A number of 7. Some other commemorations also reveal variants; this question requires future study. sticheraria combine an oktoechos section with the menaion or triodion-pentekostarion; 8. The formation of the systems of numbering is revised in Карабинов (1910, pp. 13-25).

580 581 CANTUS PLANUS 2014 POLIAKOVA — THE CLASSIFICATION OF RUSSIAN STICHERARIA

According to the other system, with its roots in Jerusalem, the Sunday preceding the Other cases of proximity in Russian manuscripts are unexpected if compared to the First Sunday of the above-mentioned system was considered to be the first Lenten Byzantine practice, as it is in case of the stichera theotokia from the fifteen antiphons Sunday (in the Studite system, Cheesefare Sunday), and the Sunday that in the Studite of Holy Friday.10 It is possible that the coincidence of the theotokia is explained by the Typikon came to be regarded as the First, that of the Triumph of Orthodoxy, was seen existence of a common source from which they were copied. as the Second, etc. Traces of this ancient numbering system appear in the majority of Russian sticheraria (Sof. 96, Typ. 148, Sof. 85, Usp. 8), in spite of the fact that, having Variety in the disposition of stichera within sequences of identical content was observed originated in the tradition of the Studite typikon, they never completely followed this by Schidlovsky (2002, pp. 117-121) in his analysis of the Fifth and Sixth weeks of Lent. system. The traces of Jerusalem may represent a characteristic of the common early He distinguished three types of content. The first is characteristic of Chil. 307, Typ. source, later corrected. Amongst the sources, the Synodal triodion and Typography 148, Sof. 85 and Sof. 96, and the second was found in Synod. 278, Usp. 8 and Typ. 147. sticherarion 147 reveal a great degree of proximity. However, the organization into groups which seems to be consistent in the analysis represented by Schidlovsky is broken when we turn, for example, to the system of In the calculations of the pentekostarion period we also see two different systems. counting the Sundays of the period of Pentecost, when the Usp. 8 and Sof. 85 disagree, Here again, the sticherarion Typ. 147 is close to Voskr. 27, the pentekostarion which or consider the title for fourth Sunday of Lent, when Usp. 8 and Sof. 85 stand by the forms a pair with Synodal triodion. Three other manuscripts mark the Sundays after same side, contrasting to the Typ. 147. Anti-Pascha as the Second, Third, Fourth and Fifth, but two manuscripts attribute the number seven to the next Sunday, of the Holy Fathers. Three manuscripts consider the Conclusion Sunday of Myrrhbearers as the third. None of them marks the Sunday of the Blind Man The seven Russian sticheraria and the triodion and pentekostarion set from the GIM – like the typikon copy prescribes – as the sixth. On the whole, there are no complete collection, which includes the sticherarion sequences, have a number of characteristics coincidences in the numbering of the Sundays of this period, but we see the same which distinguish these sources into a single group, against the background of known tendency, of correcting the numbering according to the Studite typikon, as we have Byzantine traditions. They are: the presence of the archaic stratum of the notation of seen in the case of the numbering of the Sundays of Lent. the Coislin type; the correlation of all the manuscripts with a textual redaction realized in the second half of the eleventh century; the sharing of a rare type of organization Let us now consider some aspects relating to the composition of sticheraria and the as a codex which places the sequences from the triodion cycle in a separate book; the disposition of hymns. In relation to the stichera prosomoia, this stratum is not regular rectifying of the Palestinian system of numbering of the Sundays of Lent and Pentecost and appears written in different ways in the Greek tradition; the Lenten prosomoia according to the Studite system; some rare qualities in both content and disposition of often represent a separate section, and are found less frequently written out in weekly the hymnography common to all sources, such as the fixing of the prosomoia of Saint sequences. The old Russian sticheraria agree to include for the weekdays only the Theodore within the weekly sequences, the absence of the prosomoia of Saint Joseph 9 prosomoia of Saint Theodore; the other concordance between the Russian sticheraria in the sticheraria, or the coincidences in the theotokia for the antiphons of Holy Friday. is revealed in the method of fixing the weekdays’ prosomoia of Saint Theodore within At the same time, some variants probably originating in Greek and Slavic traditions the weekly sequences. The composition of the stichera idiomela in the seven sticheraria of divergent historical backgrounds may be observed. Further research into Russian and the GIM set reflects the unabridged version of the Byzantine sticherarion (Strunk sticheraria in the circles of Greek and Slavic sources of different periods may clarify 1966, pp. 23-25). As in the case of early Byzantine manuscripts, including those such variants, permitting the continuation of their classification. with a number of correspondences, none of the sticheraria completely repeats the composition and disposition of hymns in the other. Some sequences may be identical; when there are variants, they may only concern the order of hymns in the sequences, or they may be more relevant and regard the content of the idiomela.

From the point of view of contents, some sequences in Russian sticheraria reveal an identity, as it happens in all the other Greek sources, for example, in the case of the sequences of the Second, Fourth and Fifth Sundays of Lent (Bertonière 1997, pp. 66-69).

9. Only in Synod. 319 the prosomoia by Saint Joseph are introduced, notated by a hand which differs from the main 10. As Strunk (Follieri-Strunk 1975, p. 40) noted, «no two sources call for exactly the same theotokian in exactly the same neumatic hand and which seems to be responsible for the notation of the newly introduced material. order».

582 583 CANTUS PLANUS 2014

Elizabeth J. Markham Bibliography

Bertonière Gabriel (1997), The Sundays of Lent in Triodion: The Sundays Without a Commemoration, A schematic musical form for singing Buddhist hymns Roma: Pontificio Instituto Orientale (Orientalia Christiana Analecta 253). in the ‘bilingual’ literary culture of early Japan Follieri Enrica - Strunk Oliver (1975), Triodium Athoum, Copenhagen: Munksgaard (Mmb, Série Principale 9).

Jakobson Roman ed. (1957), Fragmenta Chilandarica Palaeoslavica, Sticherarium, Hauniae: Munksgaard (Mmb 5A). Singing Chinese verse was a common improvisatory practice at the early Japanese Карабинов Иван А. (1910), Постная Триодь. Исторический обзор ея плана, состава, редакций court, and it endured against a backdrop of singing Buddhist hymns in the same и славянских переводов [Lenten triodion. The historical review of its plan, contente, redactions and Slavic translations], Saint Petersburg: V.D. Smirnov´s Typography. Chinese verse-forms in the early Japanese temple. But latest by the eleventh century, when the term rôei (literally, ‘the voiced performance of verse’) had attached itself Mares Frantisek V. (1978), Fragments du Sticherarion de Chilandar à Prague in Christian Hannick ed., Fundamental Problems of Early Slavic Music and Poetry, Copenhagen: Munksgaard (Mmb, to a particular courtly vocal repertory built on verse-fragments written in Chinese, Subsidia 6). and when ‘pointing’ of kada hymn-texts with dots of apparent tonal significance is documented, stabilisation of rules and procedures of the original improvisatory Металлов Васисий М. (1906), Богослужебное пение русской церкви в период домонгольский по историческим, археологическим и палеографическим данным [Sacred chant of Russian practice seem to have led to reliance on schematic musical form(s) for carrying Church of the pre-Mongol period acording to the historical, archaeological and paleographic Chinese verse – on a principle of standard text-setting template combined with data], 2 vols, Moscow: n.pr. modal tune-model. Contracted and re-stretched, its melodic gestures decomposed Момина Майа А. (1992), ‘Проблема правки славянских богослужебных гимнографических книг and rebuilt, such a combinatory form came to accommodate – this paper proposes на Руси в xi столетии’ [The problem of correction of the Slavic liturgical hymnographic books for the corpus it investigates – various verse-forms of varying line-lengths, Chinese in Russia in the eleventh century], Труды Отдела древнерусской литературы 35, pp. 208-215. and Japanese, sacred and secular, and for various modes of vocal rendition. Poliakova Svetlana (2009), Sin 319 and Voskr 27 and the Triodion Cycle in the Liturgical Praxis in Russia during the Studite Period, Ph. D. dissertation, Universidade Nova, Lisboa. Collating evidence for the singer’s formulaic elaboration of a tune-model from Schidlovsky Nikolas (2002), A New Folio for Mt. Athos MS Chilandari 307, with Some Observations on realisations frozen in large bodies of neumations for temple kada, extant from the the Contents of the Slavic Lenten Sticherarion and Pentekostarion in Peter Jeffery ed., The Study thirteen century, with a remarkable synoptic musical analysis of the procedure of Medieval Chant. Paths and Bridges, East and West. In Honor of Kenneth Levy, London: Boydell. for the kada-hymns, based on oral transmission but invoking conventions and Strunk Oliver (1966), Specimina Notationum Antiquiorum, Copenhagen: Munksgaard (Mmb, Pars terminology of segmental tone and poetic form, and recorded in a performance- Suppletoria 7). orientated treatise of the time, has shed light on the extent and sophistication of the – (1977), Essays on Music in the Byzantine World, New York: Norton. complex of skills, linguistic and musical, required of Buddhist cantor and courtly singer for evaluating the constraints of a shared musical structure each time anew for these various verse-forms and their linguistic and poetic conditions.

A hypothetical text-setting template Laurence Picken proposed long ago (1966, 1969) that secular court-music tunes from Táng China (618–907) preserved in Japan in instrumental tablatures in measures of eight beats would be suitable for carrying Chinese lyrics in the standard Táng verse-form, in lines of seven characters – or better for our purposes in places in this paper, in lines of seven glyphs, where ‘glyph’ refers neutrally to a unit of writing, Chinese or Japanese (example 1, p. 586).

584 585 CANTUS PLANUS 2014 MARKHAM — SINGING BUDDHIST HYMNS IN EARLY JAPAN

Example 1. Text-setting template for Chinese verse of the Táng (618-907) in lines of seven characters Example 2a. Thirteenth century kada-hymnary neumation (first hymn) in text-setting template

Text-lines would be vocalised as seven syllables, in Chinese: for each, one syllable would be set per beat on beats one to six in a measure, with the final syllable on beat seven occupying two beats. His hypothetical text-setting template was confirmed more than thirty years ago from the temple (Markham 1985): by a thirteenth-century Japanese neumation of a Buddhist eulogy vocalised as ondoku ‘reading by sound’, that is, using the Sino-Japanese pronunciations, which may be monosyllabic or polysyllabic, that approximate the sounds of the monosyllabic Chinese words the text-glyphs represent, and set to a melody found also in mensural tablature in the The text-setting template for placing a Chinese verse-form in time seems to cooperate, court-music repertory. Here with two-fold metrical expansion, lines of seven text- then, with a stabilized melody-form. The melody-form is backed up, though, by what glyphs are set to eight binary-units of two beats each (example 1): in each line, glyphs has shown up from a range of early Japanese repertories in this mode as an underlying one to six occupy one binary-unit (two beats) with the final glyph occupying two ‘global’ modal framework (example 3), a support apparently working as identified for binary-units (four beats), the individual glyphs vocalised, as stated already, either other melodic traditions as a modal memory-structure (Hopkins 1986, p. 170), and monosyllabically or polysyllabically but within the unit assigned to that glyph. Sliding that here conceives of diatonic Dorian pitch-space as divided into discrete spatial this text-setting template behind melodic neumations for Buddhist kada-quatrains in units (Markham 2012, p. 257). Slotted-in to the template, conventionalized melodic lines of seven characters (four columns of seven Chinese glyphs read top to bottom in gestures pace out the mould of a kada-hymn,5 their moving within and among these an original notation)1 likewise vocalised as ondoku in Sino-Japanese pronunciations, divisions or ‘spatial units’ controlled by dictates of the modal substructure. from a late thirteenth-century Japanese hymnary with over 200 kada,2 and then collating the slotted melodic gestures with the analysis of kada in a contemporaneous, Example 3. ‘Global’ modal memory-structure for Dorian mode-key hyôjô performance orientated treatise3 have given us some idea of the rules for singing the Chinese quatrain in medieval Japan (Markham 2008). The treatise gives a column- by-column, neume-by-neume distillation and cataloguing of the standardized set of melodic gestures available for each text-character-slot in the repertory of neumated kada – all Dorian on E and mainly pentatonic. Essentially a melodic kada-grammar it lays out (example 2a)4 in terms of invariant line-initials (thin vertical boxes) and constrained slots – constant slots (thick boxes), variable or transformational slots (thin boxes), and ‘both-and’ slots (both constant and transformational – thick dashed box) where, in our deduced metrical grid-framework, structural and ornamental dictates of For kada-hymns in the other standard Táng Chinese verse-form (the five-character the mode were evidently to be strictly followed in an elaboration of a kada tune-model. quatrain) the hymnary itself instructs in notes jotted on its reverse how the seven- character template is to be curtailed:6 by leap-frogging slots 3 and 4 in all four lines 1. Regular kada-hymns in Chinese text-form may be called shi-kada, where shi refers to verse-form. For an accessible bar the second, where slots 2 and 3 are those cut out – to retain original slot 4, facsimile of the hymn transnotated in example 2a, p. 587, see Markham (2008, figure 15).

2. Shokyô yôbun kada-shû, from facsimile publication in Shinma et al. 1984.

3. Kada-kuketsu, in Shinma et al. 1984, pp. 95-103 and pp.104-110. 5. For a tentative abstraction of a mould, see Markham (2013, p. 11).

4. Item 1 in the first scroll of the three that comprise the hymnary. See Shinma et al. (1984, p. 13 and p. 36). 6. For a fuller account of procedures of model curtailment for this verse-form, see Markham (2013, pp. 11-13).

586 587 CANTUS PLANUS 2014 MARKHAM — SINGING BUDDHIST HYMNS IN EARLY JAPAN strategically vital in modal terms for levering the melody up into the ‘mediating’ B-G The original verse is a single couplet, two lines of seven Chinese characters, seven conduit-unit, a melodic reversal needed to achieve a line-final cadence on G. Fitted glyphs. For the neumation, each line is first spliced into ‘halves’ of four and three together, and as corroborated in neumated ondoku realisations of this curtailed glyphs. For the first two of the ‘new’ four lines of the spliced text shown here as example model frozen in the thirteenth century hymnary, few slots are unaccounted for by the (example 4), on top is the original Chinese, in the middle its glyphs as neumated for the notes on the hymnary-reverse (Markham 2013, pp.12-13); but, as again documented kundoku rendition, and below and in the transliteration how it is completely vocalised in actual neumations (ibidem, pp. 14-15) and relevant for our next consideration in Japanese according to Lurie’s ‘association’ of monosyllabic Chinese words with here, ‘free’ slots in the curtailed model have proven especially useful, in one certain polysyllabic Japanese words, ‘transposition’ in Japanese word order (the sweeping context at least, for increased syllabic loading in text-setting related adaptations for arrows), and ‘addition’ of grammatical elements. carrying the Japanese waka verse-form, a five-line form in a patterned mix of both pentasyllabic and heptasyllabic units (Heldt 2008, p. 123; Markham 2013, pp. 13-16). Example 4. Kundoku: ‘association, transposition, addition’ (Lurie 2011)

The couplet stretched over the double-couplet template-cum-melody-form Courtly rôei-texts are fragments of Chinese verse, however, predominantly just single couplets. At typical two-line length of fourteen or ten characters or glyphs, they are only half the size of the four-line, double-couplet kada hymn-texts – with their twenty- eight or twenty glyphs. While double-couplet kada-hymns could be rendered in the way we have seen, as ondoku ‘reading by sound’,7 where each character is read in the order it is written and is vocalised using the Sino-Japanese pronunciation that approximates the sound of the monosyllabic Chinese word it represents, and although for the well known celebratory ‘New Year’s’ rôei, Kashin reigetsu, an ondoku reading is known and preserved in early neumations, this was not the usual mode of coping A tentative reading of the corresponding first half of the neumation of the spliced single- with verse written in Chinese, outside the temple, especially. Linguist David Lurie has couplet ‘lotus verse’ has already suggested a modus operandi with an underlying tune- recently laid out the much more common way of rendering texts written in Chinese, model that is perhaps but a variant of what we have deduced for the double-couplet kada the ‘on the spot translation’ mode of reading them in Japanese called kundoku, literally tune-model, albeit one that appears to characterize the early rôei tradition as we know ‘reading by gloss’. He explains it as ‘association, transposition, and addition’, a complex it so far (Markham 2013, p. 20). But, for the present, it is again the temple that comes to of practices that: associate logographs of Chinese origin [the individual monosyllabic our aid with a clear-cut recipe: specifically, and once more from the thirteenth-century Chinese characters] with Japanese words [typically polysyllabic] and transpose the kada-hymnary, with what are even called kun-kada, laid out in full neumations with resulting words into Japanese word order while adding necessary grammatical the ‘Japanese’ kundoku process and its dynamic state in syllable count per line, applied elements, thereby producing an actual or imagined vocalisation in Japanese.8 therein to the very model we have already examined as evaluated and neumated for its ‘Chinese’ ondoku readings of double-couplet kada. The first item of the hymnary’s Crucially for a singer (working as it seems for rôei too with a tune-and-text-setting- kun-kada section9 shows but mild effects of, and patterned responses to, the kundoku model-complex) the process produces in the main a widely varying accretion process, in terms of syllable proliferation and accommodation; each line of the kun- of syllables with consequent irregularity in line-length. As sample rôei I take a kada is indeed assigned to a separate line of the kada tune-model, accretion of syllables soteriological Buddhist ‘lotus verse’ associated with rebirth in Buddhist paradise. Its in the kundoku vocalisation means that melodic gestures in certain slots are more than brief text is a couplet by a Japanese poet inscribed in Chinese verse-form in the earliest well occupied syllable-wise, and that simple decomposition of those melodic gestures compilation of verses actually known as rôei, the early eleventh-century Wakan rôei- into constituent elements – decomposition of melodic gestures all predominantly in shû widely known in English as ‘Japanese and Chinese Poems to Sing’ (Rimer and ‘free’ slots of the model – enables their accommodating multiple syllables. Four items Chaves 1997). And it is known in an early neumation, although now in damaged on in the hymnary (Shinma et al., pp. 52, 64) however, more extreme proliferation of condition, in kundoku reading. syllables produced by the kundoku process midway in text-line three (example 2b),10

7. Lurie (2011) has «reading by voice» (or ‘sound’). 9. Item 44, in the middle scroll of the three that comprise the hymnary. See Shinma et al. (1984 p. 51 and p. 64).

8. Lurie 2011, p. 175; bracketed additions are mine. 10. Item 48, Shinma et al., p. 52 and p. 64.

588 589 CANTUS PLANUS 2014 MARKHAM — SINGING BUDDHIST HYMNS IN EARLY JAPAN requires a kun-kada both to negotiate the constraints of framing and inner constant Bibliography and variable-transformational slots in the underlying template-cum-melody-form for this line and, as well, to generate, mainly within remaining ‘free’ slots, largely syllabic Heldt Gustav (2008), The Pursuit of Harmony: Poetry and Power in Early Heian Japan, Ithaca, NY: East figurations derived from, and so compatible with, decompositions of model gestures Asia Program of Cornell University (Cornell East Asia Series 139). available for those slots. Thereby it allocates eighteen glyphs over its eight slots; with Hopkins Pandora (1986), Aural Thinking in Norway: Performance and Communication with the polysyllabic pronunciation of the two Chinese glyphs (large squares in example 2b) «Hardingfele», New York: Human Sciences Press. it includes in its neumation, when vocalised the line accommodates in total twenty Lurie David B. (2011), Realms of Literacy: Early Japan and the History of Writing, Cambridge, MA: syllables – almost three times the count for a single line in Picken’s template (example Harvard University Press. 1) for the seven-character heptasyllabic Táng verse-form, when vocalised in Chinese; Markham Elizabeth J. (1985), Tunes from Tang China at Court and Temple in Medieval Japan: First Steps and exactly the ‘Chinese’ count for an entire regular four-line kada-hymn in the five- Towards Reading Early Japanese Neumatic Notations in Hans Åstrand - Gunnar Larsson eds., character pentasyllabic Táng verse-form.11 Trends and Perspectives in Musicology: Proceedings of the World Music Conference of the International Music Council Ocober 3-5 1983, Stockholm: Royal Swedish Academy of Music, pp. 351-358.

Example 2b. Proliferation and accommodation of syllables in line three of a kun-kada – (2008), Medieval Chinese Hymns in Japanese Buddhist Liturgy: Structure and Ornament in Manuel P. Ferreira ed., Medieval Sacred Chant: From Japan to Portugal – Canto sacro medieval: do Japão a Portugal, Lisbon: Edições Colibri, pp. 11-44.

– (2012), Habitable Surface, Habitable Space: Graphic Contour and Melodic Constraints for Buddhist Chant in Early Japan in Robert Klugseder ed., Cantus Planus: Study Group of the International Musicological Society, Wien: Österreichische Akademie der Wissenschaften, Kommission für Musikforschung, pp. 250-258.

— (2013), ‘Delighting in an Art of Sino-Japanese Musical Doubleness in Premodern Japan’, The Journal of Rhythms. Official Journal of the Japan Institute of Rhythm 14, pp. 4-24.

Picken Laurence E. R. (1966), ‘Secular Chinese Songs of the Twelfth Century’, Studia Musicologica Academiae Scientiarum Hungaricae 8, pp. 125-172. Conclusion For the linguistic situation in Heian Japan, David Lurie does not like the term bilingual. — (1969), Tunes Apt for T’ang Lyrics from the Shô Part-books of Tôgaku in Korean Musicological Society ed., Essays in Ethnomusicology – A Birthday-offering for Lee Hye-ku, Seoul: Seoul National He suggests that the complex of skills involved in processing via the kundoku mode a text University Press, pp. 401-416. arranged in accordance with Chinese vocabulary and syntax would be better described Rimer J. Thomas - Chaves Jonathan (1997) eds. and trans., Japanese and Chinese Poems to Sing: The as interlingual, notwithstanding that the end-product is typically a rearrangement read Wakan Rôei Shû, New York: Columbia University Press (Translations from the Asian Classics). off in Japanese; for, as he points out, the process could also be reverse-engineered to yield from Japanese a text in Chinese-style (Lurie 2011, p. 180). The kundoku process Shinma Shin’ichi, Inui Katsumi, Fukushima Kazuo, Takahashi Shûei eds. (1984), Kanazawa bunko shiryô zenshû 7: Kayô – Shômyô-hen [Complete Materials of Kanazawa Library, Volume 7: Song – renders rôei and kun-kada identical in terms of linguistic character, overlap of genre Shômyô-section], Tôkyô: Benridô. and performance context is well established for the day, and a combinatory text-setting template cum schematic modal memory-model appears to have supported singer and cantor alike. Perhaps – it might be conjectured for the future – the principle of the schematic musical form worked in this way whether for spontaneous processing of a verse, or for a singing from memory of a song or hymn learnt off, or even for a singing supported by the songbook or hymnary? For now, though, it appears safe to say that the general system needed to rely on considerable skills of singers (and singer-scribes?) operating with their ondoku and kundoku linguistic toolboxes within shared musical structures whose constraints each instance of a verse-form evaluated in performance.

11. The issue of rhythmisation in general is deferred for the present.

590 591 International Musicological Society STUDY GROUP CANTUS PLANUS

Cantus Planus Seventeenth meeting of the IMS Study Group

Venice, Italy 28 July - 1 August 2014

with the participation of Monday 28 July

10:00-15:00 DESK AUDITORIUM—REGISTRATION

16:30-17:30

AUDITORIUM Opening Ceremony and Welcome Address

Davide Croff, Chair (Fondazione Ugo e Olga Levi, Venice) Luisa M. Zanoncelli, Scientific Board (Fondazione Ugo e Olga Levi, Venice) Dinko Fabris, Chair (IMS, Basle) Nausica Morandi, Coordinator (IMS-Study Group Cantus Planus / Fondazione Ugo e Olga Levi, Venice) James Borders, Chair (IMS-Study Group Cantus Planus)

16:30-17:30

AUDITORIUM—SESSION 1 Venice

Peter Jeffery The Choir of San Marco in a 13th-Century Mosaic

Manuel Pedro Ferreira Venetian Influence in 15th-Century Portugal

18:00 BASAGLIA ROOM—COCKTAIL RECEPTION

19:30

SAN SERVOLO CHURCH Opening Concert

Letizia Butterin, organ

21:00 DINING HALL—DINNER Tuesday 29 July 14:00-15:30 ROOM 2—POSTERS (I) Debra Lacoste Karin Strinnholm Lagergren Xaver Kainzbauer Martin Kaiser CANTUS Database The Medieval Parchment The Thesaurus Gregorianus Thesaurus Gregorianus, 8:30-10:00 Wrapper Project Database Data File: Its Musical an Internet Database of now Online Dimension Gregorian Office Antiphons: ROOM 6—SESSION 2 ROOM 7—SESSION 3 Its Textual Features and Inventories and Catalogues Old Hispanic Chant Application to the Question of the Origin of Gregorian ˯ Ana Cizmic´ Emma Hornby & Rebecca Maloy Chant Catalogue of Medieval Music Manuscripts in Dalmatia A Taxonomy of Old Hispanic Cadences: Methodology and Challenges Santiago Ruiz & Juan Pablo Rubio, OSB Liturgical Fragments of the Diocese of Sigüenza Raquel Rojo Carrillo 15:30 COFFEE BREAK (11th-16th Centuries) The Verpertini as Witnesses to the Transmission of Old Hispanic Chant in Traditions A and B Karin Strinnholm Lagergren Liturgical Music Manuscripts through Five Centuries in 16:00 TRANSFER TO SAN FRANCESCO DELLA VIGNA. THE MEETING POINT IS AT SAN SERVOLO LANDING, BOARDING AREA the Birgittine Abbey Maria Refugie (the Netherlands)

10:00 COFFEE BREAK 16:30 Visit to the Library of San Francesco della Vigna

10:30-12:00 18:30 ROOM 6—SESSION 4 ROOM 7—SESSION 5 CHURCH OF SAN FRANCESCO DELLA VIGNA CONCERT Chants for Special Occasions Cantor and Cantus — Chant from Medieval Women’s Monasteries Harald Buchinger Joseph Dyer Chants of Hildegard von Bingen, Harrad von Hohenburg, from Codex Las Huelgas and Cantus Sibyllae Chants for Exceptional Occasions and the Proper of a The Image of the Cantor in the Writings of Amalar Regular Custom: The pre-Mass Lustration in the History of Metz Ensemble Oktoechos and Prehistory of the Processional Lanfranco Menga, conductor Océane Boudeau Michelle Urberg Le sanctoral de la cathédrale de Sens (XIIe-XIVe s.): The Paschal and Mariological Processions Brothers ses particularités, son évolution et ses conséquences sur and Sisters at Vadstena Abbey le chant 19:30 TRANSFER TO SAN SERVOLO

Melanie Batoff Jean-François Goudesenne The Visitatio sepulchri as a Gospel Harmony in Political Cantus: Apostolicity and Liturgical Exports Medieval Germany in Southern France and Alps (9th-11th Centuries) 20:30 DINING HALL—DINNER

12:15 DINING HALL—LUNCH

14:00-15:30

ROOM 6—SESSION 6 ROOM 7—SESSION 7 Musical Intonations in Monastic Manuscript Collections

Sources from Italy (with a Special Silvia Tessari Focus on the Veneto) (Panel 1) The Byzantine Musical Manuscripts of the Veneto Region: Overview of a Current Research Project. One Nausica Morandi example (Bassan. gr. 34B19) The Office of Santa Cecilia in Musico-liturgical Manuscripts Rebekka Sandmeier Imposing European Culture on the Cape Colony: Cristina Bernardi Medieval Manuscripts in the Grey Collection The Feasts of the Inventio and Exaltatio Crucis in the Carthusian Antiphonary Marco Gozzi Manuscripts in Cortona: Fragments and Liturgical Books Diego Toigo in the Archivio Storico Diocesano Late Monodic Intonations of the Passion in Italy Wednesday 30 July 14:00-15:30 ROOM 2—POSTERS (II) Debra Lacoste Karin Strinnholm Lagergren Xaver Kainzbauer Martin Kaiser CANTUS Database The Medieval Parchment The Thesaurus Gregorianus Thesaurus Gregorianus, 8:30-10:00 Wrapper Project Database Data File: Its Musical an Internet Database of now Online Dimension Gregorian Office Antiphons: ROOM 6—SESSION 8 ROOM 7—SESSION 9 Its Textual Features and Sources (I) Cantus and Computing Application to the Question of the Origin of Gregorian Orsolya Csomó Geert Maessen Chant Marian Feasts in the Zagrebian Mass-Liturgy First Results of a Computational Analysis of Old in the Context of the Mediterranean Area Hispanic Chant

Anna Vildera Kate Helsen 15:30 COFFEE BREAK The Processional of Santa Maria ‘della Fava’ A New Way to See Neumes: The Optical Neume Recognition Project in Action Nuria Torres ˯ Medieval Fragments with Marian Polyphony Debra Lacoste & Jan Kolácek CANTUS for Office and Mass: Building an Online 16:00-17:00 Network of Chant Databases ROOM 6—SESSION 14a ROOM 7—SESSION 15a Heirmologion and Tropologion Classification and Systems 10:00 COFFEE BREAK Sandra Martani Svetlana Poliakova The Manuscript Cryp. E.γ.III: A 12th-Century An Hypothesis for the Classification of Russian Studite Heirmologion of the South-Italy Area Sticheraria

10:30-12:00 Svetlana Kujumdzieva Elizabeth J. Markham ROOM 7—SESSION 10 ROOM 6—SESSION 11 The Tropologion Vaticanus Graecus 771 A Schematic Musical Form for Singing Buddhist Hymns in the ‘Bilingual’ Literary Culture of Early Japan Transmission and Influence, Chants of the Mass Ordinary (I) Central Europe Marit Johanne Høye Kyrie Chants in Manuscripts from the German-speaking Hanna Zühlke 17:00-19:30 Area Hirsau in Norditalien: Zur liturgischen Musikpraxis des Benediktinerklosters Moggio im 12. und 13. ROOM 6—SESSION 14b ROOM 7—SESSION 15b Hana Vlhová-Wörner Jahrhundert Agnus Pairing and Disappearing Chant and Liturgy in Latin Southern Music, Text, Socio-political Context Réka Miklós Italy in the Middle Ages and Function in Medieval Saints’ Sarah Ann Long Der Seckauer Liber Ordinarius von ca. 1595 (Panel 2) Mensurally Notated Mass Ordinaries from Cambrai Office. New Approaches (A-Gu 1566) als letztes Dokument der mittelalterlichen and Tournai (Panel 3) Salzburger-Seckauer Liturgie und Musik Luisa Nardini The Mass of the Dead in Beneventan Manuscripts Morné Bezuidenhout In Search of the Black Swan: A Computer-aided Alejandro Enrique Planchart Approach to Interval Pattern Recognition 12:15 DINING HALL—LUNCH Melodic and Formal Syntax of the Bevenentan Proses Dirk van Betteray Thomas Forrest Kelly & Katarina Livljanic Textinterpretation und Formelkomposition - Ferial Office in Beneventan Sources of the 11th and 12th Heiligenoffizien im Codex Hartker 14:00-15:30 Centuries Roman Hankeln ROOM 7—SESSION 12 ROOM 6—SESSION 13 Matthew Peattie The Articulation of Direct Speech in High and Late Graphic Difference and the Interpretation of the Sources (II) Medieval Historiae Transmission and Influence, Climacus in Beneventan Notation (II) Susan Boynton Central Europe Danette Brink Bibiana Gattozzi Liturgy and History in the Cluniac Collection Paris, The Office in Honour of Simeon, the Recluse from Trier Gábor Kiss The Hymn in Beneventan Manuscripts BnF lat. 17716 Late reflorescence of the Alleluia Repertory in Sebastián Salvadó Central Europe Jeremy Llewellyn San Marco in Pilastro: Ranieri Zeno and the Historiae of St Mark in Venice The History and Historiography of the ‘Thomas- Katarina Šter Graduale’ (Leipzig, Universitätsbibliothek, ms. 391) Melodic History of a Music Manuscript: A Case of the Medieval Carthusian Antiphoner Jurij Snoj 20:00 DINING HALL—DINNER The Antiphoner of Isola and the Question of its Zsuzsa Czagány & Ágnes Papp Provenance Traditio Iohannis Hollandrini: Spätmittelalterliche Theorie und Choralpraxis 21:30 ROOM 6—GENERAL MEETING Thursday 31 July 16:00-17:00

ROOM 6—SESSION 20a ROOM 7—SESSION 21a 8:30-10:00 Manuscript Transmission in Italy Beneventanum

ROOM 7—SESSION 16 ROOM 6—SESSION 17 Leandra Scappaticci Sophie Burton Notation (I) Venerating the Saints (I) In the Delta of the Adriatic Sea: Pomposa Abbey Liturgical echoes: A triplet for the Exaltatio sanctae in the Guidonian Era crucis in Benevento, Biblioteca Capitolare MS 40 Franz Karl Prassl Hrvoje Beban ‘Orthographieregeln’ in der Notation adiastematischer Venerating the Founder Saint: The Office for St. Dominic James Borders St. Galler Handschriften des 10. und 11. Jahrhunderts in Croatian and Polish Sources in Comparison to the A Northern Italian Intermediary between Avignon Dominican Prototype and Rome? Oxford Bodleian Library, Canon. Lit. 375 Antanina Kalechyts and the Chants of the Pontificale Romanum Die Entwicklung der Notation als Indikator für einen Alison Altstatt Wandel der Interpretation am Beispiel des Proprium der Singing the Saints in Medieval Eichstätt Heiligen Lucia in St. Galler Handschriften vom Beginn des 10. bis zur Mitte des 11. Jahrunderts Christelle Cazaux-Kowalski 17:00-19:00 A Thirteenth-Century Officium sancti sudarii from the ROOM 6—SESSION 20b ROOM 7—SESSION 21b Cistercian Monastery of Cadouin (France, Périgord) Editing Medieval Monophonic Music: Polyphony

Current Problems and Methods Spyridon Antonopoulos 10:00 COFFEE BREAK between Old and New Philologies ‘The Organika Kratemata’ of Manuel Chrysaphes (Panel 4) Carmen Julia Gutiérrez The Role of the Canons Regular in the Dissemination 10:30 TRANSFER TO SAN MARCO. THE MEETING POINT IS AT SAN SERVOLO LANDING, BOARDING AREA Andreas Haug Towards a Semiotically Informed Transcription Practice of Medieval Polyphony in Spain

Konstantin Voigt Alexander Lingas Reconstructing Acts of Writing-Editorial Consequences Reflections on the Development of Polyphonic Singing 11:00 of the Notational Plurality of Paris 1139 in Post-Byzantine Chant MONUMENTAL ROOMS OF THE NATIONAL MARCIANA LIBRARY—VISIT TO THE EXHIBIT Cantus Planus David Catalunya Eustathios Makris The Role of Material Philology in the Process of Editing A Falsobordone Setting of Miserere in Byzantine Notation Byzantine Musical Notation in the Codices from Marciana Library Aquitanian Versus

Elaine Hild Working Realities of the New Philology 12:30 TRANSFER TO SAN SERVOLO

19:30 13:00 DINING HALL—LUNCH BASAGLIA ROOM—GALA DINNER

14:30-15:30 21:00

ROOM 7—SESSION 18 ROOM 6—SESSION 19 SAN SERVOLO CHURCH—CLOSING CONCERT Venerating the Saints (II) Tropes and Sequences Scivias. Heaven and Heart in Hildegard von Bingen Music Selected Chants of Hildegard von Bingen Tova Leigh-Choate Giulia Gabrielli Weekly Offices for Patron Saints: A Preliminary Tropes in Cantus Planus Sources from South Tyrol Ensemble Oktoechos Investigation of Their Occurrence and Composition Lanfranco Menga, conductor Arturo Tello Ruiz-Pérez Stefania Roncroffi Rethinking Partially-texted Sequence: Ecce puerpera Sources of the Office of Saint Prosper genuit, between Italy and Catalonia

15:30 COFFEE BREAK Friday 1 August

8:30-10:30

ROOM 6—SESSION 22 ROOM 7—SESSION 23 Notation (II) Byzantine and Western Connections:

Elsa De Luca Doxa in ipsistis and O quando in cruce A Taxonomy of León 8’s Notation / Ὅτε τῷ σταυρῷ in East and West (Panel 5) Laura Albiero The Development of ‘Comasca’ Notation: from France Charles M. Atkinson to Northern Italy On the Melodic Tradition of the Doxa in ipsistis

Andreas Pfisterer Gunilla Iversen Zur Bedeutung von Oxeia/Acutus/Virga in den Editing the text of Doxa en ipsistis in Latin sources griechischen und lateinischen Neumenschriften Gerda Wolfram Giovanni Varelli The Byzantine Tradition of the Great Doxology Il più antico testimone della notazione Nonantolana Nina-Maria Wanek O quando in cruce - Ὅτε τῷ σταυρῷ Revisited

10:30 COFFEE BREAK

11:00

ROOM 6 Farewell

12:15 DINING HALL—LUNCH Fondazione Ugo e Olga Levi San Servolo servizi San Marco 2893 - 30124 Venezia Isola di San Servolo - 30124 Venezia t. + 39 041 786777 t. + 39 041 2765001 [email protected] [email protected] www.fondazionelevi.it www.sanservolo.provincia.venezia.it All’indirizzo internet www.fondazionelevi.it è consultabile il catalogo delle pubblicazioni. Alcuni volumi possono essere scaricati gratuitamente in formato pdf. I volumi possono essere acquistati presso la Fondazione Ugo e Olga Levi [email protected] Ims Study Group Cantus Planus

contents Davide Croff, Presentation; Roberto Calabretto, Luisa Zanoncelli, Preface; James Borders, Foreword to the Meeting; Nausica Morandi, Opening address; Abbreviations; I - notation Essays by Andreas Pfisterer, Laura Albiero, Elsa De Luca; II - manuscripts, fragments, and sources David Catalunya, Marco Gozzi, Karin Strinnholm Lagergren, Santiago Ruiz Torres and Juan Pablo Rubio Sadia, Rebekka Sandmeier, James Borders, Jurij Snoj, Réka Miklós; III - Editing medieval monophony: between the old and new philologies Andreas Haug, Konstantin Voigt, Elaine Stratton Hild; IV - Chant and computing Debra Lacoste and Jan Koláček, Kate Helsen - Inga Behrendt - Jennifer Bain; V - Alleluias, tropes, sequences, and the mass ordinary Gábor Kiss, Giulia Gabrielli, Marit Johanne Høye, Sarah Ann Long, Hana Vlhová-Wörner; VI - Saints offices: new themes and contexts Morné Bezuidenhout, Dirk van Betteray, Roman Hankeln, Danette Brink; VII - Regional and local traditions Raquel Rojo Carrillo, Hanna Zühlke, Nausica Morandi, Stefania Roncroffi, Diego Toigo, Jean-François Goudesenne, Océane Boudeau, Tova Leigh-Choate, Melanie Batoff, Alison Altstatt, Michelle Urberg, Manuel Pedro Ferreira, Orsolya Csomó; VIII - Chant and liturgy in latin Southern Italy Thomas Forrest Kelly and Katarina Livljanić, Luisa Nardini, Matthew Peattie, Bibiana Gattozzi; IX - Byzantium and the East Svetlana Kujumdzieva, Silvia Tessari, Gerda Wolfram, Nina-Maria Wanek, Svetlana Poliakova, Elizabeth J. Markham.

ISBN 9788875520588

9 788875 520588 >